Novel polyoxometalate silica nano-sized spheres: Efficient catalysts for olefin oxidation and the deep desulfurization process

Article abstract of DOI:10.1039/c3dt53444h

A novel method to prepare silica nano-sized particles incorporating polyoxometalates was developed leading to a new efficient heterogeneous oxidative catalyst. Zinc-substituted polyoxotungstate [PW₁₁Zn(H ₂O)O₃₉]^5- (PW₁₁Zn) was encapsulated into silica nanoparticles using a cross-linked organic-inorganic core, performed through successive spontaneous reactions in water. The potassium salt of PW₁₁Zn and the composite formed, PW₁₁Zn-APTES@SiO ₂, were characterized by a myriad of solid-state methods such as FT-IR, FT-Raman, ³¹P and ¹³C CP/MAS solid-state NMR, elemental analysis and SEM-EDS, confirming the integrity of the PW ₁₁Zn structure immobilized in the silica nanoparticles. The new composite has shown to be a versatile catalyst for the oxidation of olefins and also to catalyze the desulfurization of a model oil using H₂O ₂ as the oxidant and acetonitrile as the solvent. The novel composite material was capable of being recycled without significant loss of activity and maintaining its structural stability for consecutive desulfurization and olefin oxidative cycles.

Full text of DOI:10.1039/c3dt53444h

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Novel polyoxometalate silica nano-sized spheres:
efficient catalysts for olefin oxidation and the deep
desulfurization process†
Cite this: Dalton Trans., 2014, 43,
9518
Lucie S. Nogueira,a Susana Ribeiro,a Carlos M. Granadeiro,a Eulália Pereira,a
Gabriel Feio,b Luís Cunha-Silvaa and Salete S. Balula*a
A novel method to prepare silica nano-sized particles incorporating polyoxometalates was developed
leading to a new efficient heterogeneous oxidative catalyst. Zinc-substituted polyoxotungstate [PW11Zn-
5−
(H2O)O39
]
(PW11Zn) was encapsulated into silica nanoparticles using a cross-linked organic–inorganic
core, performed through successive spontaneous reactions in water. The potassium salt of PW11Zn and
the composite formed, PW11Zn-APTES@SiO2, were characterized by a myriad of solid-state methods
such as FT-IR, FT-Raman, 31P and 13C CP/MAS solid-state NMR, elemental analysis and SEM-EDS, con-
firming the integrity of the PW11Zn structure immobilized in the silica nanoparticles. The new composite
has shown to be a versatile catalyst for the oxidation of olefins and also to catalyze the desulfurization of
a model oil using H2O2 as the oxidant and acetonitrile as the solvent. The novel composite material was
capable of being recycled without significant loss of activity and maintaining its structural stability for con-
secutive desulfurization and olefin oxidative cycles.
Received 7th December 2013,
Accepted 3rd April 2014
DOI: 10.1039/c3dt53444h
www.rsc.org/dalton
economical and environmentally attractive oxidation catalysts
in both research and industrial processes.5 Hydrogen peroxide
Introduction
The design of novel heterogeneous catalysts easily recyclable is one of the most attractive oxidants, mainly because it is
without a loss of activity and selectivity is a challenging goal of environmentally clean and easily handled.6 Recently, remark-
liquid phase oxidation catalysis. The development of eco- able interest has been noticed for the application of hetero-
sustainable catalytic systems to transform cheap compounds geneous catalyst based POMs in the oxidation of various
into valuable intermediates for organic synthesis is an area of valuable molecules (alkenes, alkanes, alcohols and sulfides) in
interest for both the laboratory and industry. Polyoxometalates the presence of H2O2, mainly due to the necessity of recovering
(POMs) are a class of well-defined early transition metal and recycling these active catalytic compounds. Different
oxygen clusters, which exhibit distinctive structures and methodologies have been studied using silica as a solid
various functionalities.1 POMs have received increasing atten- support to immobilize POMs via dative, covalent or electro-
tion as oxidative catalysts due to a unique combination of pro- static binding.1,7–17 One of the most widely used immobili-
perties, including thermal and oxidative stability, tuneable zation strategies is electrostatic bonding to NH2-modified
acidity, redox potentials, solubility, etc.1–3 Transition-metal mesoporous silica or to imidazolium ionic liquid immobilized
mono-substituted polyoxotungstates contain an active center on silica. In both cases the POM is linked via anion exchange
M isolated in the tungsten oxide matrix and strongly bound or formation of ion pairs.17,18 Furthermore, some hetero-
through M–O–W bridges, which may confer an additional cata- geneous POM based silica catalysts have been prepared by
lytic activity for oxidative reactions and a high capacity as immobilization of POMs through electrostatic interaction with
oxygen transfer agents.2,4 These compounds have shown to be the surface of positively charged silica nanoparticles.19–21
Another well-known strategy is the encapsulation of POMs into
core/silica nanoparticles using a reverse micelle and sol–gel
technique.22–25 In particular, the sol–gel method is a promis-
ing approach to heterogenize POMs through occlusion of the
POM into an inert matrix. Immobilization of POMs by surfac-
tant encapsulation has also been recently explored to form
supramolecular hybrid catalysts by replacement of the counter
cations from the POMs by quaternary ammonium cations,
aREQUIMTE & Departamento de Química e Bioquímica, Faculdade de Ciências,
Universidade do Porto, 4169-007 Porto, Portugal. E-mail: sbalula@fc.up.pt;
Tel: +351 220 402 576
bCENIMAT/I3N, Departamento de Ciência dos Materiais, Faculdade de Ciências e
Tecnologia, Universidade Nova de Lisboa, 2829-516 Monte da Caparica, Portugal
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c3dt53444h
9518 | Dalton Trans., 2014, 43, 9518–9528
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
such
as
di-(11-hydroxyundecyl)dimethylammonium equipped with a Nd:YAG laser with an excitation wavelength of
(DOHDA).26,27 In general, diffusion restriction of the POM 1064 nm and the laser power set to 350 mW. Electronic absorp-
within the silica matrix seems to be crucial to prevent POM tion spectra were acquired on a Varian Cary 50 Bio spectrophoto-
leaching into solution.
meter. 31P NMR spectra were recorded on a Bruker Avance III
In a sequence of our recent studies concerning the prepa- 400 using CD3CN as the solvent and the chemical shifts are
ration of heterogeneous catalysts by incorporation of POMs given relative to 85% H3PO4 as an external standard. Solid-
into distinct solid supports,15–17,28–30 we proposed a new state 31P MAS NMR spectra were acquired with a 7 T (300 MHz)
methodology to prepare POM based silica composites in situ in AVANCE III Bruker spectrometer under a magic angle spinning
a cross-linked organic/inorganic hybrid core in an effort to of 10 kHz at room temperature. The spectra were obtained by a
establish a facile strategy to prepare uniform and well dis- solid echo sequence with an echo delay of 15 μs, a 90° pulse of
persed POMs inside silica spheres. The catalytic core is formed 10.5 μs at a power of 20 W, and a relaxation delay of 30 s.
by the mono-substituted [PW11Zn(H2O)O39]5− (PW11Zn) linked Potassium phosphate (K3PO4) was used as a reference. Scan-
to the amine-organosilane (APTES) surrounded by a silica shell ning electron microscopy (SEM) and energy dispersive X-ray
(Fig. S1 in ESI†). The resulting composite, PW11Zn-APTES@ spectroscopy (EDS) studies were performed at “Centro de Mate-
SiO2, is expected to combine the advantages of molecular com- riais da Universidade do Porto” (CEMUP, Porto, Portugal)
plexes and reusable solids. The PW11Zn was judiciously using a scanning electron microscope JEOL JSM 6301F operat-
chosen because the application of this POM as a catalyst for ing at 15 kV equipped with an energy-dispersive X-ray spectro-
oxidative systems remains practically unexplored. There is only meter Oxford INCA Energy 350. The samples were studied as
one published example of the use of PW11Zn as a homo- powders and were subjected to gold sputtering before analysis.
geneous catalyst for the oxidation of alcohols31 and, to our The catalytic reactions were monitored by GC-FID performed
knowledge, only one other report describes the immobilization in a Bruker 430-GC. Hydrogen was used as the carrier gas
of PW11Zn in a silica support material, although anchored on (55 cm3 s−1) and fused silica Supelco capillary columns SPB-5
the surface of amorphous silica.32 The composite particles of (30 m × 0.25 mm i.d.; 25 μm film thickness) were used.
PW11Zn-APTES@SiO2 were tested as a heterogeneous catalyst
Synthesis and preparation of materials
in two different catalytic systems using H2O2 as an oxidant.
The oxidation and epoxidation of olefins originate oxygenated
Zinc-substituted polyoxotungstate. The potassium salt of
products of great interest.33 The second catalytic system con- [PW11Zn(H2O)O39]5− (PW11Zn) was prepared according to a
sists in its application in oxidative desulfurization systems previously described procedure.38 The POM was characterized
(ODS). In fact, the development of efficient ODS systems is by UV-Vis spectroscopy, vibrational spectroscopy (FT-IR and
crucial for the production of ultra low levels of sulfur fuels FT-Raman) and 31P NMR spectroscopy that allowed to
and, in general, POMs have shown to be efficient catalysts for positively confirm the identity of the compound.
oxidative desulfurization.29,30,34–36 However, the application of
PW11Zn-APTES@SiO2 nanospheres. The silica nanoparticles
solution containing APTES
POM based silica composites as heterogeneous catalysts has were prepared by adding
a
been poorly explored for the ODS technology.36,37
(0.64 mmol) to a solution of PW11Zn (0.054 mmol) in water
(10 mL). The mixture was then kept under stirring at
room temperature for 2 h. Then, ethanol (8.75 mL), TEOS
(1.2 mmol) and ammonia (1.5 mL) were slowly added under
stirring to the initial suspension. After stirring for 24 h at
room temperature, the mixture was centrifuged and the silica
Experimental section
Materials and methods
All the reagents used in the preparation of the POMs and the composites were washed thoroughly with an ethanol–water
silica composites, namely sodium tungstate dihydrate 1 : 1 solution and dried in a desiccator over silica gel. PW11Zn-
(Aldrich), sodium hydrogen phosphate dihydrate (Aldrich), APTES@SiO2: Anal. found (%): W, 20.1; C, 9.3; N, 2.6; loading
zinc acetate dihydrate (May & Baker), nitric acid 65% (Merck), of PW11Zn: 0.099 mmol per 1 g and ratio of APTES/PW11Zn =
potassium chloride (Aldrich), ethanol (Aga), tetraethoxysilane 19. Selected FT-IR (cm−1): 3412, 2925, 1644, 1338, 1132, 1044,
(TEOS, Aldrich), (3-aminopropyl)triethoxysilane (APTES, 944, 904, 798, 770, 690, 668, 618, 584, 574, 560, 542, 438.
Aldrich) and ammonia 25% (Merck), were used as received. Selected FT-Raman (cm−1): 3302, 2918, 1607, 1454, 1412, 1310,
Geraniol 98% (Aldrich), cis-cyclooctene 95% (Aldrich), styrene 1231, 1142, 1048, 978, 959, 857, 754, 587, 500, 434.
98% (Aldrich), dibenzothiophene 99% (Aldrich), 4,6-dimethyl-
Catalytic olefin oxidations. The oxidation reactions of cis-
dibenzothiophene 98% (Aldrich), acetonitrile (MeCN, cyclooctene (i), geraniol (iii) and styrene (v) (Scheme 1) were
Panreac), and hydrogen peroxide 30% (Riedel-de-Häen) were carried out in MeCN, in a borosilicate 5 mL reaction vessel,
purchased from commercial suppliers and used without with addition of aqueous H2O2 (30 wt%), in the presence of
further purification. Elemental analyses were performed by potassium salt of the zinc-substituted POM (PW11Zn) or the
ICP-MS on a Varian 820-MS at the University of Santiago de composite PW11Zn-APTES@SiO2. The oxidation of geraniol
Compostela. Infrared absorption spectra were recorded on a was studied at room temperature and protected from light,
Jasco 460 91 Plus using KBr pellets, while the FT-Raman whereas the oxidative reactions of cis-cyclooctene and styrene
spectra were recorded on a RFS-100 Bruker FT-spectrometer were studied at 70 °C. In a typical experiment, the catalyst
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 9518–9528 | 9519

View Article Online
Paper
Dalton Transactions
model oil and MeCN. In a typical catalytic reaction, 30 mg of
PW11Zn-APTES@SiO2 (containing the equivalent of 3 µmol of
PW11Zn) were added to 0.75 mL of MeCN and 0.75 mL of the
model oil. This mixture was stirred for 10 min until the initial
extraction equilibrium was reached. An aliquot from the upper
oil phase was taken. The catalytic reaction was initiated by the
addition of the oxidant H2O2 (75 µL). The DBT and the 4,6-
DMDBT contents in the model oil were quantified periodically
by GC analysis using tetradecane as a standard. At the end of
each ODS process the heterogeneous catalyst PW11Zn-APTES@
SiO2 was recovered by filtration, washed several times with
MeCN, dried at room temperature overnight and then reused
in a new ODS cycle under the same reaction conditions as
before.
Results and discussion
Catalyst characterization
The PW11Zn-APTES@SiO2 composite was prepared through
the initial formation of a cross-linked organic/inorganic hybrid
core and the subsequent formation of a silica shell by the alka-
line hydrolysis of TEOS. The POM-containing silica spheres
were characterized by several characterization techniques
including vibrational spectroscopy (FT-IR and FT-Raman),
solid-state 31P and 13C CP/MAS NMR spectroscopy, elemental
analysis, scanning electron microscopy (SEM) and energy dis-
persive X-ray spectroscopy (EDS). The vibrational spectroscopy
data of the composite material are shown in Fig. 1. The FT-IR
spectrum of PW11Zn-APTES@SiO2 exhibits the main bands
associated with the silica framework stretches, namely the
intense band located at 1044 cm−1 assigned to the νas(Si–O–Si)
stretch, as well as the νas(Si-OH), νs(Si–O–Si) and δ(Si–O–Si)
stretching modes located at 944, 798 and 438 cm−1, respecti-
vely.39,40 These intense bands mask some of the bands associ-
ated with the zinc phosphotungstate vibrational modes.
Nevertheless, comparing the spectrum of the composite with
the typical bands of silica frameworks, it is possible to observe
the presence of two bands located at 904 (sh.) and 770 cm−1
that should correspond to the νas(W–Ob–O) and νas(W–Oc–O)
stretches of the POM, respectively.39,41 The presence of the
POM in the composite is more evident in the FT-Raman spec-
trum since the intense silica bands that overlap and mask the
POM bands in FT-IR are less active in Raman.24 The FT-Raman
spectrum of the PW11Zn-APTES@SiO2 is mainly composed
by some of the typical POM stretching modes, namely the
νas(P–Oa), νas(W–Od), νs(W–Od) and νas(W–Ob–O) vibrations
Scheme 1 Chemical structure of the substrates investigated and their
respective oxidation products.
(3 μmol of PW11Zn or 30 mg of PW11Zn-APTES@SiO2) was
added to 1 mmol of the substrate in MeCN (1.5 mL) under stir-
ring. The reaction was started by addition of H2O2 to the reac-
tion mixture, 4.5 mmol for the oxidation of (iii) and (v) and
1 mmol for the oxidation of (i). The reactions were followed by
GC analysis and stopped when a complete conversion of the
substrate was observed or when the product yields remained
constant after two successive GC analyses. At regular intervals,
an aliquot was taken directly from the reaction mixture with a
microsyringe, diluted in MeCN, centrifuged (when necessary)
and injected into the GC or GC-MS equipment for analysis of
starting materials and products. When the composite catalyst
was used, it was filtered off at the end of reactions, and
washed with MeCN several times to remove the remaining sub-
strate, reaction products and the oxidant. The recovered cata-
lyst was dried at room temperature overnight and reused in a
new reaction under identical experimental conditions, with
readjustment of all quantities, without changing the molar
ratios and reaction concentrations. Blank reactions were per-
formed for all substrates, confirming that no oxidation pro-
ducts are obtained unless the catalyst and H2O2 are present.
Catalytic oxidative desulfurization process (ODS). The oxi-
dative desulfurization studies were performed using a model
oil containing dibenzothiophene (DBT, ix in Scheme 1) and
4,6-dimethyldibenzothiophene (4,6-DMDBT, x in Scheme 1) in
n-octane as the representative refractory sulfur-compounds in
diesel. The ODS experiments were carried out under air
(atmospheric pressure) in a closed borosilicate 5 mL reaction
vessel equipped with a magnetic stirrer and immersed in a
thermostatically controlled liquid paraffin bath at 50 °C. The
ODS studies were performed using the heterogeneous catalyst
PW11Zn-APTES@SiO2 in a biphasic system formed by the
located at 1048, 978, 959 and 857 cm−1, respectively.15,41
A
small shift to lower wavelengths relative to the isolated POM is
observed, especially in the bands assigned to the ν(W–Od)
stretches, probably due to the interaction between the POM
and the support as previously reported.17,42 The FT-Raman
spectrum also exhibits some bands assigned to the APTES
modes, including a broad band located at approximately
2918 cm−1 assigned to the νas(C–H) and νs(C–H) stretching,
the bands at 1454 and 1412 cm−1 arising from the CH2 scissor-
9520 | Dalton Trans., 2014, 43, 9518–9528
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 2 31P MAS NMR spectra of PW11Zn, and the silica-based composite
before, PW11Zn-APTES@SiO2, and after catalysis, PW11Zn-APTES@SiO2-ac.
PW11Zn and PW11Zn-APTES@SiO2 composites. The spectrum
of PW11Zn exhibits a main peak at −14.55 ppm and two
smaller peaks at −13.25 and −16.02 ppm. The main peak is
assigned to the PW11Zn units while the smaller ones may
result from different degrees of hydration of the same units
since the 31P MAS NMR signals of heteropolyanions are known
to be very sensitive to the hydration effect.45,46 A fourth peak is
also observed in the spectrum (δ = −11.40 ppm) with an even
smaller intensity than the previous ones and that should
correspond to a small amount of [PW11O39]7− anions formed
during the synthesis of PW11Zn.15,47
After the incorporation into the silica matrix, the 31P MAS
NMR studies indicate that PW11Zn is distributed over two
different environments, one with the two typical peaks at
−11.55 and −13.00 ppm (similar to free PW11Zn) and a second
environment with a broad peak centred at −1.05 ppm. The
latter peak is typical of silica-supported Keggin-type phospho-
metalate anions.32,45,46,48 The down-field shift relative to iso-
lated POMs is a result of the deshielding effect on the
phosphorus atoms.46 The broadness of the peak is due to the
Fig. 1 FT-IR (A) and FT-Raman (B) of PW11Zn, and the POM-containing
silica nanospheres before, PW11Zn-APTES@SiO2, and after catalysis (ac),
PW11Zn-APTES@SiO2-ac.
ing and CN stretch, respectively, and a band at 1310 cm−1 different orientations of PW11Zn units within the silica matrix,
from the ν(C–C) stretching.43,44 The vibrational spectroscopy leading to slightly different environments around the phos-
results obtained for the PW11Zn-APTES@SiO2 composite are in phorus atoms and a larger dispersion of the 31P signal. More-
good agreement with the values described in the literature for over, integration of 31P signals shows that the support-
transition-metal substituted POMs immobilized in silica interacting PW11Zn units are the predominant species in the
matrices.17,32
composite material.
The amount of PW11Zn per gram of composite
The PW11Zn-APTES@SiO2 composite was also studied by
(0.099 mmol) was quantified by elementary analysis. The 13C CP/MAS NMR spectroscopy. The spectrum of the compo-
amount of APTES was also calculated to be 1.9 mmol, which site before catalytic use (Fig. 3) exhibits three peaks located at
indicates the presence of a large excess of APTES compared 43, 22 and 10 ppm. The peaks are assigned to the C3, C2 and
with PW11Zn in the hybrid core of silica particles (ratio APTES/ C1 carbon atoms, respectively, of the aminopropyl group,49
PW11Zn = 19). The presence and structural integrity of the and indicate that these groups remain intact in the final com-
POM in the silica matrix were also evaluated by 31P solid-state posite material.50 Moreover, the absence of 13C signals from
NMR spectroscopy. Fig. 2 shows the 31P MAS NMR spectra of the ethoxy carbons (ca. 18 and 60 ppm) points to the practi-
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 9518–9528 | 9521

View Article Online
Paper
Dalton Transactions
Fig. 5 Conversion data obtained for the oxidation of different sub-
strates after 24 h of reaction using the homogeneous PW11Zn and the
heterogeneous PW11Zn-APTES@SiO2 catalysts.
Fig. 3 13C CP/MAS NMR spectra of the PW11Zn-APTES@SiO2 before
and after catalysis (ac); the inset shows the chemical structure of APTES
with carbon numbering of the aminopropyl group.
Catalytic olefin oxidation
The oxidation of cyclooctene (i), geraniol (iii) and styrene (v)
was performed using the heterogeneous catalyst PW11Zn-
APTES@SiO2 that was prepared by the immobilization of the
active center PW11Zn using a cross-linked organic–inorganic
core (the organosilane APTES and the POM PW11Zn). The cata-
lytic studies were carried out with H2O2 as the oxidant and
MeCN as the solvent. In the absence of a catalyst there was no
conversion for the oxidation of the substrates studied.
The three distinct substrates were selected to evaluate the
catalytic performance of the new composite PW11Zn-APTES@
SiO2 and its efficiency was compared with the homogeneous
catalyst PW11Zn. Fig. 5 displays the conversion data obtained
for the oxidation of the various substrates after 24 h. It can be
observed that the heterogeneous PW11Zn-APTES@SiO2 pre-
sents a similar or even higher catalytic activity than the homo-
geneous PW11Zn. Furthermore, the kinetic profiles of the
homogeneous and heterogeneous catalysts were compared and
are presented in Fig. S2 of ESI† for the oxidation of cyclooctene
and in Fig. 6 for the oxidation of geraniol and styrene. A com-
plete conversion was achieved in the presence of the compo-
site after 24 h of reaction for the oxidation of styrene and
geraniol. It is possible to observe that for geraniol oxidation
the composite shows a better activity from the first minutes of
reaction. In fact, after 10 min the conversion for geraniol oxi-
Fig. 4 SEM images of PW11Zn-APTES@SiO2 (a) before and (c) after cata-
lysis and the corresponding EDS spectra (b, d).
cally complete hydrolysis and/or condensation reactions of dation was 72% in the presence of the composite and 27% in
APTES.50,51 The 13C CP/MAS results are in good agreement the presence of the homogeneous PW11Zn. Only after 6 h of
with the values reported in the literature for APTES in siliceous reaction, similar conversion data were achieved using the
supports.52
PW11Zn and the composite (93% for PW11Zn and 99% for the
The morphology and chemical composition of the PW11Zn- composite). The kinetic profiles for styrene oxidation (Fig. 6)
APTES@SiO2 were studied by scanning electron microscopy demonstrate that the catalytic activity between PW11Zn and the
(SEM) and energy dispersive X-ray spectroscopy (EDS). SEM composite PW11Zn-APTES@SiO2 is similar in the first 6 h of
images (Fig. 4) show that the composite material is composed reaction and a significant superior conversion was achieved
of uniformly dispersed silica nanospheres with approximately using the heterogeneous catalyst after 24 h (100% for the
350 nm diameter, and EDS confirms the presence of PW11Zn composite and 77% using the homogeneous PW11Zn). The
in the silica nanospheres. In addition to the intense peak of increased catalytic activity found for the composite compared
silicon, the EDS spectrum clearly indicates the presence of with the homogeneous PW11Zn in the presence of an excess of
phosphorus, tungsten and zinc that constitute the POM.
the H2O2 oxidant is probably due to the higher stability of the
9522 | Dalton Trans., 2014, 43, 9518–9528
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 1 The selectivity of the various products obtained from styrene
oxidation catalyzed by PW11Zn and PW11Zn-APTES@SiO2 after 24 h of
reaction
Selectivity (%)
vi
vii
viii
Conversion (%)
20
PW11Zn
80
0
20
PW11Zn-APTES@SiO2
1st cycle
2nd cycle
54
75
71
31
11
21
15
14
8
100
67
96
3rd cycle
induction periods of several hours were found. Higher catalytic
performance was obtained using iron-substituted phospho-
tungstates encapsulated into silica nanoparticles, and in this
case 96% of conversion was achieved after 3 h of reaction;
however, three different products were produced.25 Most of the
work that has been published about the oxidation of geraniol
catalyzed by Keggin type POMs was for homogeneous systems.
For metal mono-substituted phosphotungstates and borotung-
state selective oxidation of geraniol to 2,3-epoxygeraniol was
reported under similar experimental conditions to those used
in this work.14,25,55 However, in the presence of the PW11Zn-
Fig. 6 Comparison of kinetic profiles for the oxidation of geraniol (top) APTES@SiO2 catalyst, better catalytic performance and higher
and styrene (bottom), between the homogeneous (open symbols) and
heterogeneous (solid symbols) catalysts, using H2O2 as the oxidant and
MeCN as the solvent.
selectivity were obtained for less reaction time.
For styrene oxidation catalyzed by the homogeneous
PW11Zn or the composite PW11Zn-APTES@SiO2, the only
product formed in the first 6 h of reaction was the benz-
aldehyde (vi in Scheme 1); however, after 24 h, 2-hydroxy-1-
phenylethanone (viii) was also formed in the presence of both
catalysts. In addition, using the composite, benzoic acid (vii)
was also formed in a minor amount. Table 1 shows the
product distribution and the catalytic performance obtained
for styrene oxidation after 24 h of reaction in the presence of
the homogeneous and heterogeneous catalysts. The oxidation
of styrene to benzaldehyde with H2O2 catalyzed by polyoxo-
tungstates is already well documented in the literature.56–59
The mechanism proceeds initially by the interaction of H2O2
with the POM, generating active species, which may be hydro-
peroxo or bridging peroxo species. Then, styrene binds to one
of the metal–peroxo bonds to produce a peroxometallocycle,
and in the next step styrene oxide is formed. A further nucleo-
philic attack of H2O2 on the styrene oxide originates benz-
aldehyde.56,58 The formation of benzoic acid from oxidation of
benzaldehyde is well documented.60 Recently, our research
group also observed the formation of 2-hydroxy-1-phenyl-
ethanone under similar experimental conditions, and it was
confirmed that it originated from the oxidation of styrene
oxide. The formation of styrene oxide was not observed in the
present work probably due to its fast oxidation in the presence
of an excess of H2O2.61 Some reports can be found in the litera-
ture for the oxidation of styrene using heterogeneous catalysts
based on POMs immobilized in solid supports.11,56,62–65 Hu
et al. reported the preparation of heterogeneous catalysts
obtained through the impregnation of transition metal-mono-
active center PW11Zn promoted by its encapsulation in silica
nanoparticles. For the oxidation of cyclooctene an equal molar
amount of substrate/oxidant was used, and in this case a
similar activity between homogeneous and heterogeneous cata-
lysts was only achieved after 24 h (79% for the composite and
74% for PW11Zn, Fig. S2 in ESI†) and only 1,2-epoxycyclo-
octane (ii in Scheme 1) was formed. The stability of the homo-
geneous PW11Zn at the end of the reaction was investigated by
31P NMR, and two single peaks were found at 4.3 and
−10.5 ppm (Fig. S3 in ESI†). The peak at −10.5 ppm corres-
ponds to PW11Zn in CD3CN solution as presented in Fig. S3†
and the peak at 4.3 ppm could be identified as the Venturello
complex by comparison with literature data.53,54 This last
peroxocomplex must be the catalytically active species.
Geraniol (iii) is an allylic alcohol that offers several possible
positions for oxidative attack, namely at the two double bonds,
at the allylic carbon centers and at the carbon of the CH2OH
group. However, the oxidation of this substrate catalyzed by
PW11Zn and the composite with H2O2 at room temperature
only forms the 2,3-epoxygeraniol (iv in Scheme 1). Few recent
studies have been published reporting the epoxidation of gera-
niol using similar experimental conditions to this work, by
various heterogeneous catalysts based on POMs immobilized
onto mesoporous silica materials functionalized with amine
groups.15,17 In these cases, complete epoxidation of geraniol
was achieved only after 24 h of reaction for the best catalysts,
and in the presence of mono-substituted silicotungstates,
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 9518–9528 | 9523

View Article Online
Paper
Dalton Transactions
Fig. 8 Desulfurization of substrates showing the initial extraction stage
(before the dashed line) and the catalytic stage (after the dashed line),
at 50 °C.
Fig. 7 Activity data obtained for the oxidation of geraniol and styrene
after 24 h of reaction when catalysed by the heterogeneous PW11Zn-
APTES@SiO2 for three consecutive cycles.
substituted Keggin-type POMs (XMW11O39; X = P, Si and M = been reached, the second step corresponds to the catalytic
Co, Ni, Cu, Mn) in Schiff-base modified SBA-15.56 The oxi- stage and it is started by the addition of the oxidant H2O2. In
dation of styrene was also performed with H2O2 as the oxidant this stage, the sulfur refractory compounds (DBT and 4,6-
and MeCN as the solvent, although the highest conversion DMDBT) were oxidized into sulfoxides and/or sulfones. These
value was only 57% after 12 h of reaction, significantly lower products remain in the MeCN phase since they are much more
than the catalytic results presented in this work. Patel and co- soluble in this organic polar phase than in the model oil. A
workers described the catalytic oxidation of styrene in the pres- blank reaction was performed using the same model oil and
ence of zirconia-supported POMs, in which, despite a complete the same experimental conditions but in the absence of the
conversion of styrene, the necessary reaction times are much catalyst PW11Zn-APTES@SiO2. In this case no oxidation of DBT
longer than those found in the present work.
or 4,6-DMDBT was detected.
The recyclability of PW11Zn-APTES@SiO2 was studied for
Fig. 8 shows the results obtained for the desulfurization
the cases where the heterogeneous catalyst showed a similar or process of DBT and 4,6-DMDBT. For DBT the main extraction
even higher performance than the homogeneous PW11Zn, i.e. occurred in the initial extraction step, before addition of H2O2.
geraniol and styrene oxidation. The composite was recovered After 10 min of stirring at 50 °C, 84% of the initial DBT in the
from the reaction mixture by centrifugation, washed several model oil was extracted into the MeCN phase. A lower extrac-
times with MeCN and dried at room temperature to be used in tion yield (38%) was observed for 4,6-DMDBT. This distinct be-
a fresh reaction under identical experimental conditions havior between the two sulfur compounds can be caused by
(using the same ratio of catalyst : substrate : oxidant : solvent). the higher solubility of the DBT in the MeCN and also by the
Fig. 7 shows the reusability of the composite for the oxidation steric hindrance of the methyl groups in 4 and 6 positions of
of both substrates. For the oxidation of geraniol no loss of 4,6-DMDBT that is expected to have a slower diffusion rate and
activity was noticed between the consecutive cycles. For styrene thus a slower extraction.
oxidation, a small decrease of activity was noticed from the
After the addition of H2O2, the oxidation of DBT and 4,6-
first to the second cycle; however, the activity is then enhanced DMDBT occurred in the presence of PW11Zn-APTES@SiO2,
for the third cycle. Some differences of selectivity for the probably in the MeCN phase since sulfones or sulfoxides were
various products formed could also be observed between the not detected in the oil phase. The consumption of non-oxi-
different catalytic cycles (Table 1). Higher selectivity for benz- dized DBT and 4,6-DMDBT in the extracting solvent leads to a
aldehyde is achieved during the second and the third cycle.
continuous transfer of these compounds from the oil phase to
the extracting phase. The desulfurization curves displayed in
Fig. 8 demonstrate that after the addition of the oxidant, there
is an induction period in the first 2 h. This behavior must be
Oxidative desulfurization process (ODS)
The efficiency of PW11Zn-APTES@SiO2 as a heterogeneous
catalyst for the oxidative desulfurization process (ODS) was related with the mechanism of the ODS process, possibly with
investigated using a model oil containing dibenzothiophene the formation of active catalytic species by interaction of the
oxidant with the terminal bonds WVIvO or with the substi-
(DBT, ix in Scheme 1) and 4,6-dimethldibenzothiophene (4,6-
DMDBT, x in Scheme 1). These ODS studies were performed in tuted Zn–OH2 (water behaves as a labile ligand). This first step
a biphasic system formed by equal amounts of the model oil generates peroxo species that can then oxidize the sulfur com-
pounds into sulfones and/or sulfoxides.66–68 After 2 h, the
and MeCN used as an extracting solvent. The ODS process of
the model oil occurred in two distinct stages. The first corres- efficiency of the ODS process catalyzed by PW11Zn-APTES@
ponds to the initial extraction of the DBT and 4,6-DMDBT SiO2 increases and a fast desulfurization is observed for DBT
and 4,6-DMDBT. In fact, after 3 h, the model oil contained
from the oil to the solvent by stirring the biphasic system for
10 min at 50 °C. After the initial extraction equilibrium has only 10 ppm of DBT and 56 ppm of 4,6-DMDBT. The oxidation
9524 | Dalton Trans., 2014, 43, 9518–9528
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
after 2 h, instead of the 3 h necessary during the first ODS
cycle (Fig. 9).
Fig. 10 presents the desulfurization of 4,6-DMDBT for
different ODS cycles at various reaction times. Also for this
sulfur compound, the induction period observed during the
first ODS cycle is not observed for the remaining cycles.
Higher desulfurization is observed for 1, 2 and 3 h of reaction
during the second and the third cycle, indicating that the cata-
lyst is more active after the first cycle. This is probably
explained by some modification of the initial composite into a
more active catalytic composite during the induction period of
the first cycle. Apparently, this new active form is stable to
washing and drying of the catalyst, so no further induction
period occurs in subsequent cycles.
Fig. 9 Kinetic profile for DBT desulfurization for three consecutive ODS
cycles, at 50 °C.
Catalyst material stability
The integrity and structural stability of the PW11Zn-APTES@
SiO2 composite were evaluated by several techniques. The cata-
lyst was recovered after an ODS catalytic cycle (PW11Zn-
APTES@SiO2-ac, where ac stands for after catalysis) and was
studied by vibrational spectroscopy (FT-IR and FT-Raman),
elemental analyses, 31P MAS NMR and SEM/EDS. The
vibrational spectroscopy data of the composite after catalysis
(Fig. 1) still exhibit the main bands associated with the
PW11Zn stretching modes together with the bands assigned to
the silica framework (in FT-IR) and APTES (in FT-Raman). A
decrease in the intensity of the bands associated with the POM
stretches is observed, which should be due to some leaching
that may have occurred. This is more evident in the FT-Raman
spectrum, where the relative intensity of the peaks associated
with νas(W–Od) and νs(W–Od) vibrations is reduced after cata-
lytic use. The elemental analysis results of PW11Zn-APTES@
SiO2-ac indicate a POM loading of 0.093 mmol per 1 g of
material. Such a value represents a leaching of only 6% of
PW11Zn after the ODS cycle. To investigate the identity of the
leached species, the 31P NMR analysis was performed using
the resulting solution after catalytic use. This liquid NMR spec-
trum is presented in Fig. S3 in ESI† and no phosphorus signal
was detected. Furthermore, the catalytic activity of the leached
species was investigated by removing the solid catalyst after
the first hour of reaction and following the ODS system until
4 h. Fig. 9 shows that no catalytic desulfurization was achieved
in the absence of the solid. This result indicates that
0.17 µmol of the leached species are not catalytically active
and do not present phosphorus in their structure.
Fig. 10 Desulfurization data obtained for 4,6-DMDBT for three con-
secutive ODS cycles, at 50 °C.
of this last sulfur compound seems to be more difficult, poss-
ibly because its extraction from the model oil is slower than
for DBT. This is corroborated by previous reports of a lower
reactivity for the oxidation of 4,6-DMDBT than DBT. The differ-
ence in reactivity has been explained by the steric hindrance
caused by the methyl groups in 4,6-DMDBT, hindering the
interaction of this sulfur compound with active catalytic
species and the oxidant.69–75 Complete desulfurization of DBT
was achieved after 4 h, while 4,6-DMDBT is near totally
removed from oil after 5 h (Fig. 8).
The recyclability of the PW11Zn-APTES@SiO2 was investi-
gated for three consecutive ODS cycles. The solid catalyst was
recovered after each ODS cycle by simple filtration followed by
washing with MeCN, drying at room temperature, and then
reutilization in a new ODS system under the same experi-
mental conditions. Fig. 9 displays the kinetic profile of DBT
desulfurization for the three consecutive cycles. It is possible
to observe that the highest extraction of DBT from oil in con-
secutive cycles occurred during the initial extraction step.
However, comparing the catalytic stage of the three ODS
cycles, it is possible to notice that the induction period is only
observed during the first cycle and a complete desulfurization
of DBT is rapidly achieved for the consecutive cycles. In fact,
the kinetic profile for DBT desulfurization is similar for the
consecutive cycles and complete extraction of DBT is achieved
Analyzing the solid catalyst PW11Zn-APTES@SiO2-ac after
4 h of ODS use by 31P MAS NMR (Fig. 2), the spectrum exhibits
a single broad peak centred at 0.04 ppm. After catalysis, the
PW11Zn molecules seem to be distributed in a single chemical
environment. As previously discussed, this environment
should correspond to PW11Zn units interacting with the inner
silica surface. The 31P MAS NMR results suggest that the
PW11Zn fragments that were in more central positions and
experiencing less interactions with silica (hence with similar
chemical shifts to the “free” PW11Zn) have been displaced to
positions closer to the inner silica interface after the ODS
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 9518–9528 | 9525

View Article Online
Paper
Dalton Transactions
cycle. The 13C CP/MAS NMR spectrum of the composite after Dr Isabel Gonçalves from CICECO Laboratory, University of
catalytic use (Fig. 3) presents three main peaks assigned to the Aveiro, Portugal.
C3 (43 ppm), C2 (21 ppm) and C1 (9 ppm) carbon atoms of the
aminopropyl group. The chemical shifts are nearly identical
with the ones before catalysis, suggesting that the structure
of the aminopropyl groups of APTES remains essentially
unaffected after catalytic use. The smaller peaks observed in
References
the spectrum are most likely due to the presence of solvent
molecules (MeCN) along with DBT and 4,6-DMDBT oxidation
products. The integrity of the composite was also evaluated by
SEM/EDS (Fig. 4). The SEM images of PW11Zn-APTES@SiO2-ac
show that no significant changes in the morphology were
observed after catalysis. The EDS spectra before and after cata-
lysis display identical profiles, indicating the existence of
P, W and Zn from the POM, in addition to silicon from the
silica matrix and APTES.
1 M. T. Pope and A. Müller, Polyoxometalate Chemistry From
Topology via Self-Assembly to Applications, Kluwer Academic,
Dordrecht, 2002.
2 R. Neumann, in Modern Oxidation Methods, ed. J.-E. Back-
vall, Wiley-VCH, Weinheim, 2004, p. 223.
3 O. A. Kholdeeva, in Heterogeneous Catalysis Research Pro-
gress, ed. M. B. Gunther, Nova Science Publ., New York,
2008, p. 267.
4 I. V. Kozhevnikov, Chem. Rev., 1998, 98, 171–198.
5 C. L. Hill, J. Mol. Catal. A: Chem., 2007, 262, 2–6.
6 C. W. Jones, Applications of hydrogen peroxide and deriva-
tives, Royal Society of Chemistry, Cambridge, 1999.
7 A. C. Estrada, I. C. M. S. Santos, M. M. Q. Simões,
M. G. P. M. S. Neves, J. A. S. Cavaleiro and
A. M. V. Cavaleiro, Appl. Catal., A, 2011, 392, 28–35.
8 J. Kasai, Y. Nakagawa, S. Uchida, K. Yamaguchi and
N. Mizuno, Chem. – Eur. J., 2006, 12, 4176–4184.
9 O. A. Kholdeeva, N. V. Maksimchuk and G. M. Maksimov,
Catal. Today, 2010, 157, 107–113.
Conclusions
Novel silica nanoparticles incorporating [PW11Zn(H2O)O39]5−
were successfully prepared by the initial formation of a cross-
linked organic–inorganic core composed of the polyoxometa-
late and an amine-organosilane, followed by the formation of
a silica shell through the hydrolysis of TEOS. The obtained
PW11Zn-APTES@SiO2 was characterized by several techniques,
namely vibrational spectroscopy (FT-IR and FT-Raman), 31P 10 S. K. Jana, Y. Kubota and T. Tatsumi, J. Catal., 2008, 255,
MAS NMR spectroscopy, SEM and EDS. The results show that 40–47.
the material is formed by uniform silica nanoscale spheres 11 P. A. Shringarpure and A. Patel, Dalton Trans., 2010, 39,
with approximately 350 nm in diameter and containing the 2615–2621.
POM/APTES hybrid dispersed in the silica matrix. The nano- 12 R. Zhang and C. Yang, J. Mater. Chem., 2008, 18, 2691–
composite proved to be a versatile and robust heterogeneous 2703.
catalyst for the oxidation of various olefins and also for the oxi- 13 Y. Guo, Y. Yang, C. Hu, C. Guo, E. Wang, Y. Zou and
dative desulfurization of a model oil, using H2O2 as the S. Feng, J. Mater. Chem., 2002, 12, 3046–3052.
oxidant and MeCN as the solvent. The comparison between 14 S. S. Balula, I. C. M. S. Santos, L. Cunha-Silva,
the catalytic performances of the heterogeneous and homo-
geneous catalysts shows that, in general, the incorporation of
the POM into the silica spheres has led to an enhancement of
A. P. Carvalho, J. Pires, C. Freire, J. A. S. Cavaleiro, B. de
Castro and A. M. V. Cavaleiro, Catal. Today, 2013, 203, 95–
102.
the catalytic activity. The stability and robustness of the hetero- 15 C. M. Granadeiro, A. D. S. Barbosa, P. Silva, F. A. Almeida
geneous catalyst could be confirmed by different catalytic tests
Paz, V. K. Saini, J. Fires, B. de Castro, S. S. Balula and
and several characterization techniques. Following the promis-
L. Cunha-Silva, Appl. Catal., A, 2013, 453, 316–326.
ing results obtained in this work, other novel POM-supported 16 S. S. Balula, C. M. Granadeiro, A. D. S. Barbosa,
silica composites will be prepared by the method herein
described and tested in catalytic oxidations.
I. C. M. S. Santos and L. Cunha-Silva, Catal. Today, 2013,
210, 142–148.
17 S. S. Balula, L. Cunha-Silva, I. C. M. S. Santos,
A. C. Estrada, A. C. Fernandes, J. A. S. Cavaleiro, J. Pires,
C. Freire and A. M. V. Cavaleiro, New J. Chem., 2013, 37,
2341–2350.
Acknowledgements
The authors acknowledge the Fundação para a Ciência e a 18 V. N. Panchenko, I. Borbath, M. N. Timofeeva and
Tecnologia (FCT, MEC, Portugal) for their general financial S. Gobolos, J. Mol. Catal. A: Chem., 2010, 319, 119–125.
support through the strategic projects Pest-C/EQB/LA0006/ 19 N. M. Okun, T. M. Anderson and C. L. Hill, J. Am. Chem.
2011 (to Associated Laboratory REQUIMTE) and PEst-C/CTM/ Soc., 2003, 125, 3194–3195.
LA0025/2022 (Strategic Project – LA 25-2001/2012), the R&D 20 N. M. Okun, T. M. Anderson and C. L. Hill, J. Mol. Catal. A:
project PTDC/EQU-EQU/121677/2010, the fellowship SFRH/ Chem., 2003, 197, 283–290.
BPD/73191/2010 (to CG), the Portuguese Nuclear Magnetic 21 N. M. Okun, M. D. Ritorto, T. M. Anderson, R. P. Apkarian
Resonance Network (PRNMR) and the research group of
and C. L. Hill, Chem. Mater., 2004, 16, 2551–2558.
9526 | Dalton Trans., 2014, 43, 9518–9528
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
22 C. S. Neves, C. M. Granadeiro, L. Cunha-Silva, D. Ananias, 44 R. S. Pryce and L. L. Hench, J. Mater. Chem., 2004, 14,
S. Gago, G. Feio, P. A. Carvalho, P. Eaton, S. S. Balula and
E. Pereira, Eur. J. Inorg. Chem., 2013, 2877–2886.
23 M. S. S. Balula, H. I. S. Nogueira and A. M. V. Cavaleiro, in
2303–2310.
45 A. Ghanbari-Siahkali, A. Philippou, J. Dwyer and
M. W. Anderson, Appl. Catal., A, 2000, 192, 57–69.
Advanced Materials Forum Iii, Pts
P. M. Vilarinho, 2006, vol. 514–516, pp. 1206–1210.
24 C. M. Granadeiro, R. A. S. Ferreira, P. C. R. Soares-Santos, 47 G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange,
1
and 2, ed. 46 T.-h. Chang, J. Chem. Soc., Faraday Trans., 1995, 91, 375–
379.
L. D. Carlos, T. Trindade and H. I. S. Nogueira, J. Mater.
Chem., 2010, 20, 3313–3318.
J. Dutour, S. Surblé and I. Margiolaki, Science, 2005, 309,
2040–2042.
25 J. L. C. Sousa, I. C. M. S. Santos, M. M. Q. Simões, 48 F. Lefebvre, J. Chem. Soc., Chem. Commun., 1992, 756–757.
J. A. S. Cavaleiro, H. I. S. Nogueira and A. M. V. Cavaleiro, 49 M. R. Mello, D. Phanon, G. Q. Silveira, P. L. Llewellyn and
Catal. Commun., 2011, 12, 459–463.
26 Y. Zhao, Y. Li, W. Li, Y. Wu and L. Wu, Langmuir, 2010, 26,
18430–18436.
27 W. Qi, Y. Z. Wang, W. Li and L. X. Wu, Chem. – Eur. J.,
2010, 16, 1068–1078.
28 S. S. Balula, I. C. M. S. Santos, L. Cunha-Silva,
C. M. Ronconi, Microporous Mesoporous Mater., 2011, 143,
174–179.
50 Z. Luan, J. A. Fournier, J. B. Wooten and D. E. Miser, Micro-
porous Mesoporous Mater., 2005, 83, 150–158.
51 A. Bordoloi, N. T. Mathew, F. Lefebvre and S. B. Halligudi,
Microporous Mesoporous Mater., 2008, 115, 345–355.
A. P. Carvalho, J. Pires, C. Freire, J. A. S. Cavaleiro, B. de 52 P. D. Young and J. M. Notestein, ChemSusChem, 2011, 4,
Castro and A. M. V. Cavaleiro, Catal. Today, 2013, 203, 95–
102.
29 S. Ribeiro, A. D. S. Barbosa, A. C. Gomes, M. Pillinger,
I. S. Goncalves, L. Cunha-Silva and S. S. Balula, Fuel
Process. Technol., 2013, 116, 350–357.
1671–1678.
53 C. Aubry, G. Chottard, N. Platzer, J. M. Bregeault,
R. Thouvenot, F. Chauveau, C. Huet and H. Ledon, Inorg.
Chem., 1991, 30, 4409–4415.
54 A. C. Dengel, W. P. Griffith and B. C. Parkin, J. Chem. Soc.,
Dalton Trans., 1993, 2683–2688.
30 S. Ribeiro, C. M. Granadeiro, P. Silva, F. A. Almeida Paz,
F. F. de Biani, L. Cunha-Silva and S. S. Balula, Catal. Sci. 55 I. C. M. S. Santos, M. M. Q. Simões, M. M. M. S. Pereira,
Technol., 2013, 3, 2404–2414.
R. R. L. Martins, M. G. P. M. S. Neves, J. A. S. Cavaleiro and
A. M. V. Cavaleiro, J. Mol. Catal. A: Chem., 2003, 195, 253–
262.
31 Z. Nadealian, V. Mirkhani, B. Yadollahi, M. Moghadam,
S. Tangestaninejad and I. Mohammadpoor-Baltork,
J. Coord. Chem., 2012, 65, 1071–1081.
32 B. J. S. Johnson and A. Stein, Inorg. Chem., 2001, 40, 801–
808.
56 J. Hu, K. Li, W. Li, F. Ma and Y. Guo, Appl. Catal., A, 2009,
364, 211–220.
57 E. Poli, R. De Sousa, F. Jerome, Y. Pouilloux and
J.-M. Clacens, Catal. Sci. Technol., 2012, 2, 910–914.
33 J. M. Derfe and M. M. Derfe, in Encyclopedia of Chemical
Technology Kirk-Orthmer, ed. S. Maurice, Wiley, New York, 58 P. Shringarpure and A. Patel, React. Kinet. Mech. Catal.,
1978, vol. 22. 2011, 103, 165–180.
34 Y. Chen, S. Zhao and Y.-F. Song, Appl. Catal., A, 2013, 466, 59 J. Tang, X.-L. Yang, X.-W. Zhang, M. Wang and C.-D. Wu,
307–314. Dalton Trans., 2010, 39, 3396–3399.
35 Y. Liu, S. Liu, S. Liu, D. Liang, S. Li, Q. Tang, X. Wang, 60 M. R. Maurya, P. Saini, C. Haldar and F. Avecilla, Poly-
J. Miao, Z. Shi and Z. Zheng, ChemCatChem, 2013, 5, 3086–
3091.
36 X.-M. Yan, P. Mei, L. Xiong, L. Gao, Q. Yang and L. Gong,
Catal. Sci. Technol., 2013, 3, 1985–1992.
hedron, 2012, 31, 710–720.
61 C. M. Granadeiro, P. Silva, V. K. Saini, F. A. A. Paz, J. Pires,
L. Cunha-Silva and S. S. Balula, Catal. Today, 2013,
218–219, 35–42.
37 Z. E. A. Abdalla and B. Li, Chem. Eng. J., 2012, 200–202, 62 P. A. Shringarpure and A. Patel, Chem. Eng. J., 2011, 173,
113–121. 612–619.
38 N. Mizuno, K. Yamaguchi and K. Kamata, Coord. Chem. 63 S. Pathan and A. Patel, Dalton Trans., 2011, 40, 348–355.
Rev., 2005, 249, 1944–1956.
39 X. Luo and C. Yang, Phys. Chem. Chem. Phys., 2011, 13,
7892–7902.
40 R. C. Schroden, C. F. Blanford, B. J. Melde, B. J. S. Johnson
and A. Stein, Chem. Mater., 2001, 13, 1074–1081.
41 C. Rocchiccioli-Deltcheff, M. Fournier, R. Franck and
R. Thouvenot, Inorg. Chem., 1983, 22, 207–216.
42 Y. Guo, C. Hu, C. Jiang, Y. Yang, S. Jiang, X. Li and
E. Wang, J. Catal., 2003, 217, 141–151.
64 N. K. K. Raj, S. S. Deshpande, R. H. Ingle, T. Raja and
P. Manikandan, Catal. Lett., 2004, 98, 217–223.
65 X. Yu, L. Xu, X. Yang, Y. Guo, K. Li, J. Hu, W. Li, F. Ma and
Y. Guo, Appl. Surf. Sci., 2008, 254, 4444–4451.
66 H. Li, X. Jiang, W. Zhu, J. Lu, H. Shu and Y. Yan, Ind. Eng.
Chem. Res., 2009, 48, 9034–9039.
67 Y. Ding, W. Zhu, H. Li, W. Jiang, M. Zhang, Y. Duan and
Y. Chang, Green Chem., 2011, 13, 1210–1216.
68 X. Xue, W. Zhao, B. Ma and Y. Ding, Catal. Commun., 2012,
29, 73–76.
43 M. Hiraoui, M. Guendouz, N. Lorrain, A. Moadhen, L. Haji
and M. Oueslati, Mater. Chem. Phys., 2011, 128, 151– 69 C. Komintarachat and W. Trakarnpruk, Ind. Eng. Chem.
156.
Res., 2006, 45, 1853–1856.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 9518–9528 | 9527

View Article Online
Paper
Dalton Transactions
70 M. Te, C. Fairbridge and Z. Ring, Appl. Catal., A, 2001, 219, 73 W. Zhu, W. Huang, H. Li, M. Zhang, W. Jiang, G. Chen and
267–280. C. Han, Fuel Process. Technol., 2011, 92, 1842–1848.
71 X. Jiang, H. Li, W. Zhu, L. He, H. Shu and J. Lu, Fuel, 2009, 74 J. H. Xu, S. Zhao, W. Chen, M. Wang and Y. F. Song, Chem.
88, 431–436. – Eur. J., 2012, 18, 4775–4781.
72 M. A. Rezvani, A. F. Shojaie and M. H. Loghmani, Catal. 75 W. Trakarnpruk and K. Rujiraworawut, Fuel Process.
Commun., 2012, 25, 36–40.
Technol., 2009, 90, 411–414.
9528 | Dalton Trans., 2014, 43, 9518–9528
This journal is © The Royal Society of Chemistry 2014