Phosphotungstates as catalysts for monoterpenes oxidation: Homo- and heterogeneous performance

Article abstract of DOI:10.1016/j.cattod.2012.02.020

Lacunary (PW₁₁), mono-substituted (PW₁₁M) and sandwich type M₄(PW₉)₂ phosphotungstates (M = Co^II, Fe^III) exhibits high catalytic activity for geraniol and R-(+)-limonene oxidation under mild conditions, using hydrogen peroxide as oxidant and acetonitrile as solvent. The best homogeneous performance was found for the PW₁₁ and for the Co₄(PW₉)₂. Complete conversion of geraniol and R-(+)-limonene was achieved after 1.5 h and 3 h of reaction, respectively, when catalyzed by PW₁₁. The yield of 2,3-epoxygeraniol was 100% in the presence of lacunary PW₁₁ and mono-substituted compounds. PW₁₁ and Co₄(PW ₉)₂ were immobilized on an amine-functionalized SBA-15 (aptesSBA-15) and the composites materials were characterized by FT-IR, FT-Raman, UV-vis/DRS, ³¹P MAS NMR, elemental analysis and SEM-EDS. Their heterogeneous catalytic activity was investigated for the same substrates and compared with that in homogeneous conditions. The PW₁₁@aptesSBA- 15 and Co₄(PW₉)₂@aptesSBA-15 showed to be active heterogeneous catalysts for geraniol epoxidation with 100% of selectivity for 2,3-epoxide. R-(+)-limonene was also oxidized by PW₁₁@aptesSBA- 15 achieving 90% of conversion after 24 h of reaction. Both composites could be separated after the reaction and reused for at least three cycles without loss of activity. The stability of the heterogeneous catalysts after catalytic reactions was investigated by different techniques of characterization.

Full text of DOI:10.1016/j.cattod.2012.02.020

Catalysis Today 203 (2013) 95–102
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Phosphotungstates as catalysts for monoterpenes oxidation: Homo- and
heterogeneous performance
a,∗
b
a
c
c
Salete S. Balula , Isabel C.M.S. Santos , Luís Cunha-Silva , Ana P. Carvalho , João Pires ,
a
b
a
d
Cristina Freire , José A.S. Cavaleiro , Baltazar de Castro , Ana M.V. Cavaleiro
REQUIMTE, Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade do Porto, 4169-007 Porto, Portugal
Departamento de Química, QOPNA, Universidade de Aveiro, 3810-193 Aveiro, Portugal
Departamento de Química e Bioquímica, CQB, Faculdade de Ciências da Universidade de Lisboa, Campo Grande C8, 1749-016 Lisboa, Portugal
Departamento de Química, CICECO, Universidade de Aveiro, 3810-193 Aveiro, Portugal
a
b
c
d
a r t i c l e i n f o
a b s t r a c t
Lacunary (PW11), mono-substituted (PW11M) and sandwich type M (PW ) phosphotungstates (M = CoII,
Article history:
4
9 2
Received 9 November 2011
Received in revised form 20 January 2012
Accepted 5 February 2012
Available online 13 March 2012
III
Fe ) exhibits high catalytic activity for geraniol and R-(+)-limonene oxidation under mild conditions,
using hydrogen peroxide as oxidant and acetonitrile as solvent. The best homogeneous performance
was found for the PW11 and for the Co4(PW9)2. Complete conversion of geraniol and R-(+)-limonene
was achieved after 1.5 h and 3 h of reaction, respectively, when catalyzed by PW11. The yield of 2,3-
epoxygeraniol was 100% in the presence of lacunary PW11 and mono-substituted compounds. PW11 and
Co4(PW9)2 were immobilized on an amine-functionalized SBA-15 (aptesSBA-15) and the composites
Keywords:
Phosphotungstates
Oxidation
Hydrogen peroxide
Catalysis
3
1
materials were characterized by FT-IR, FT-Raman, UV–vis/DRS, P MAS NMR, elemental analysis and
SEM-EDS. Their heterogeneous catalytic activity was investigated for the same substrates and compared
with that in homogeneous conditions. The PW11@aptesSBA-15 and Co4(PW9)2@aptesSBA-15 showed to
be active heterogeneous catalysts for geraniol epoxidation with 100% of selectivity for 2,3-epoxide. R-
(+)-limonene was also oxidized by PW11@aptesSBA-15 achieving 90% of conversion after 24 h of reaction.
Both composites could be separated after the reaction and reused for at least three cycles without loss of
activity. The stability of the heterogeneous catalysts after catalytic reactions was investigated by different
techniques of characterization.
©
2012 Elsevier B.V. All rights reserved.
1
. Introduction
they can be seen as having reactive low valent transition metal
centers complexed by inorganic oxometalate ligands, that have
high capacity as oxygen transfer agents [2–4]. These compounds
were shown to be economically and environmentally attractive
oxidation catalysts in both academic and industrial processes
[5]. The hydrogen peroxide is the second most attractive oxi-
dant after dioxygen because it is inexpensive, environmentally
clean and easily handled [6]. During the last decade we have
shown that Keggin-type polyoxotungstates can be effective homo-
geneous catalysts in the oxidation of several alkenes and alkanes
The development of catalysts for selective and environmentally
friendly oxidation of organic compounds is nowadays a remark-
able research goal. The development of eco-sustainable catalytic
systems to transform cheap natural compounds into valuable inter-
mediates for organic synthesis is an area of interest for both in
the laboratory and in industry. The oxygenated products derived
from terpene are of great interest due to their utilization in the
production of fragrances and perfumes [1]. Generally important
prerequisites for clean and efficient technologies are stable cata-
lysts with high product selectivities and high activity at ambient
temperatures without waste of the applied oxidant. A successful
possibility is to use inorganic and therefore robust transition metal
complexes as catalysts in combination with hydrogen peroxide
as an oxygen donor. Transition metal substituted polyoxometa-
lates (POMs) are attractive catalysts for oxidative reactions because
with hydrogen peroxide [7–18]. These compounds include mono-
m−
substituted [PW11M(H O)O39]
(PW11M) and sandwich-type
2
n−
[M (H O) (PW O ) ] (M (PW ) ), where M represents several
4
2
2
9
34
2
4
II
9 2
II
III
II
II
transition metals such as Co , Fe , Mn , Cu , Ni and others. In
this paper we report the oxidation of geraniol and R-(+)-limonene
7−
with H O in acetonitrile, using the lacunary [PW11O37
]
(PW11),
2
III
2
II
the Fe and Co mono-substituted and sandwich-type polyox-
otungstates as catalysts (Fig. 1). The outcome of the reactions
depended strongly on the catalyst. To our knowledge, catalytic
studies of R-(+)-limonene oxidation using mono-substituted or
sandwich-type phosphotungstates have not been reported before.
∗ Corresponding author. Tel.: +351 220402576; fax: +351 220402659.
E-mail address: sbalula@fc.up.pt (S.S. Balula).
0
920-5861/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2012.02.020

9
6
S.S. Balula et al. / Catalysis Today 203 (2013) 95–102
Fig. 1. Schematic representation of the structures of the selected POMs: lacunar
PW11 (top left), mono-substituted PW11M (bottom left) and sandwich M4(PW9)2
Scheme 1.
(
right side).
Porto, Portugal), using a high resolution environmental scanning
electron microscope (FEI Quanta 400 FEG ESEM) equipped with an
energy-dispersive X-ray spectrometer (EDAX Genesis X4M). The
samples were analyzed as powders coated with a thin gold film.
Few papers can be found reporting the oxidation of this substrate
catalyzed by POMs and these are homogeneous and biphasic sys-
tems also using other less green oxidants [19–21]. Also, a small
number of reports on the oxidation of geraniol catalyzed by POMs
have appeared in the literature [8,15,22].
2.2. Synthesis of phosphotungstates
The high catalytic performance found for the homogeneous
studied phosphotungstates required the need of catalyst recyclabil-
ity. Thus, new heterogeneous catalysts based in phosphotungstates
were prepared by immobilization of PW11 and Co (PW ) onto
The tetrabutylammonium ([(C4H9)4N], TBA) salts of the
following compounds were prepared by known procedures:
(TBA)4H3[PW11O39] (PW11) [18], (TBA)4H[PW11Co(H2O)O39]
4
9 2
amine functionalized SBA-15 silica material. A few studies are
reporting the immobilization of polyoxotungstates on SBA-15 for
oxidative catalysis application [23–29]. To our knowledge, the
present work reports for the first time the oxidation of geraniol and
R-(+)-limonene catalyzed by heterogeneous phosphotungstates
immobilized on functionalized SBA-15 support.
(PW11Co)
[18,30],
(TBA)4[PW11Fe(H2O)O39]
(PW11Fe)
[31],
[18], (TBA)7H3[Co4(H2O)2(PW9O34)2]
(Co4(PW9)2)
(TBA)6[Fe4(H2O)2(PW9O34)2] (Fe4(PW9)2) [32,33]. All phos-
photungstates were characterized by elemental and thermal
analysis, infrared absorption spectroscopy and electronic absorp-
tion spectroscopy. Characterization of these compounds has been
compared and confirmed with published results [18,31,33].
2
. Experimental
2.3. Supported phosphotungstates
2.1. Materials and methods
Some of phosphotungstates anions prepared previously were
immobilized in large porous SBA-15. This support material
was obtained following a procedure adapted from the litera-
ture, [34] and surface of SBA-15 was modified by a grafting
methodology adapted from a previous work [35]. The dried
SBA-15 support was refluxed for 24 h in dry toluene with (3-
aminopropyl)triethoxysilane (aptes) under argon. The resulting
aptes-functionalized SBA-15 (aptesSBA-15) contained 1.2 mmol of
NH2 per 1 g of support material (4.6% of C, 1.6% of N and 1.6%
of H). Immobilization of selected phosphotungstates was per-
formed by stirring the mixture of 0.3 g aptesSBA-15 in 10 mL of
acetonitrile with 0.1 mmol of TBA salt of PW11, and Co (PW ) ,
Geraniol (Aldrich), R-(+)-limonene (Aldrich), acetonitrile (Pan-
reac), H O 30% (Riedel de-Häen, Aldrich) were used as received
2
2
without further purification.
GC–MS analysis were performed using a Hewlett Packard 5890
chromatograph equipped with a Mass Selective Detector MSD
series II using helium as the carrier gas (35 cm/s); GC-FID was
carried out in a Varian Star 3400CX chromatograph to moni-
tories catalytic reactions and a Varian CP-3380 to follow the
heterogeneous reactions. In both experiments, the helium was the
carrier gas (55 cm/s) and fused silica Supelco capillary columns
SPB-5 (30 m × 0.25 mm i.d.; 25 ␮m film thickness) were used.
Elemental analysis for W, Fe and Co were performed by ICP
spectrometry (University of Aveiro, Central Laboratory of Anal-
ysis) and C, H, N elemental analysis was performed on a Leco
CHNS-932. Hydration water contents were estimated by ther-
mogravimetric analysis on a TGA-50 Shimadzu thermobalance.
Infrared absorption spectra were obtained on a Mattson 7000 FT-IR
spectrometer, using KBr pellets. FT-Raman spectra were recorded
on a RFS-100 Bruker FT-spectrometer, equipped with a Nd:YAG
laser with excitation wavelength of 1064 nm, with laser power set
to 200 mW. Diffuse reflectance spectra (UV–vis/DRS) were regis-
tered on JascoV-560 spectrophotometer, using MgO as reference.
4
9 2
for 24 h at room temperature. The supported POM material
PW11@aptesSBA-15 and Co (PW ) @aptesSBA-15, respectively)
(
4
9 2
were filtered and washed with acetonitrile until the filtrate did
not reveal any POM (existence of POM in filtrate was followed
by UV–vis spectroscopic experiments), and were dried under vac-
◦
uum for several hours at 80 C. These composites materials were
characterized by elemental analysis (W, Co), UV–vis/DRS, FT-IR
and FT-Raman spectroscopy. The loading of phosphotungstates
achieved was 0.034 mmol for PW11@aptesSBA-15 (6.9 wt.% of W)
and 0.020 mmol for Co (PW ) @aptesSBA-15 (4.9 wt.% of W).
4
9 2
3
1
Solid-state P MAS NMR spectra were recorded on a 9.4 T Bruker
Avance 400 spectrometer. Scanning electron microscopy (SEM)
and energy-dispersive X-ray spectroscopy (EDS) studies were per-
formed at “Centro de Materiais da Universidade do Porto” (CEMUP,
2.4. Oxidation reactions
The oxidation reactions of geraniol (i, Scheme 1) and R-(+)-
limonene (iv, Scheme 1) were carried out in acetonitrile, under

S.S. Balula et al. / Catalysis Today 203 (2013) 95–102
97
air (atmospheric pressure) in a closed borosilicate 10 mL reac-
tion vessel equipped with a magnetic stirrer and immersed in a
thermostated oil bath. The oxidant used was H O2 (30 wt.%) and
2
III
II
the lacunary PW11, the mono-substituted PW M (M = Fe , Co ),
11
II
III
and the sandwich type M (PW ) (M = Fe , Co ) were tested as
4
9 2
homogeneous catalysts. In the oxidation of i the reaction was
done at room temperature and protected from light, and the
◦
oxidative reactions of iv were performed at 80 C. In a typical
experiment, the substrate (1 mmol) and the catalyst (3 ␮mol),
placed in the vessel, were dissolved in acetonitrile (1.5 ml), stirred,
and the hydrogen peroxide was added to the reaction mixture,
5
00 ␮l (4.5 mmol). The reactions were followed by GC analysis
and stopped when a completed conversion of the substrate was
observed or when the product yields remained constant after two
successive GC analyses. An aliquot was taken directly from the
reaction mixture with a microsyringe at regular intervals, diluted
in acetonitrile and injected into the GC or GC–MS equipment for
analysis of starting materials and products. The reaction products
reported here were identified as described elsewhere [36,37]. The
same procedure was performed for the heterogeneous reactions
using the supported phosphotungstates (PW11@aptesSBA-15 and
Co (PW ) @aptesSBA-15). In this case, the heterogeneous catalysts
Fig. 2. FT-IR spectra of (a) aptesSBA-15, (b) PW11@aptesSBA-15, (c)
−
1
P2Co4@aptesSBA-15, in the wavenumber region between 300 and 1900 cm
.
Spectroscopic methods including UV–vis/DRS, FT-IR, FT-Raman
and 31P MAS NMR were performed to characterize the composite
materials, in particular to verify the presence of phosphotungstate
and also the stability of these compounds after their immo-
bilization. Fig. 2 presents the FT-IR spectra of the supported
phosphotungstates and the support aptesSBA-15, in the wavenum-
4
9 2
were filtered off at the end of reactions, washed with acetonitrile
several times, dried in air at room temperature overnight and then
reused using the same reaction conditions.
−
1
ber region between 300 and 1900 cm . The support aptesSBA-15
shows the main string bands from silica framework in the region
2.5. Titration of hydrogen peroxide
−
1
of 400–1100 cm : ꢀas(Si
O Si), ꢀs(Si O Si) and ı(O Si O) [41].
The total concentration of unused hydrogen peroxide was deter-
These bands are very intense and obscure partially the charac-
teristic bands of the POMs that appear also in the same region:
mined by titration using the ceric sulphate method [38]. At regular
intervals, an aliquot was accurately weighted, dissolved in diluted
sulphuric acid and the peroxides titrated against 0.1 N ceric sul-
phate solution, using ferroin as indicador.
ꢀas(W
O W), ꢀas(W O) and ␯as(P O) [31,42–44]. Thus, compar-
ing the spectrum of the support aptesSBA-15 with spectra of
PW11@aptesSBA-15 and Co (PW ) @aptesSBA-15 it is possible to
4
9 2
−
1
−1
find two extra small bands near 880 cm and 720 cm , which may
be attributed to the ꢀas(W W) indicating the presence of phos-
O
−
1
3
. Results
photungstates. In the region of 1500–1550 cm can be observed
two weak bands corresponding to ꢀs(NH ) and ı(NH ), that con-
2
2
3
.1. Catalysts preparation and characterization
firms the incorporation of amine groups on the material [45].
The FT-Raman spectra of the supported phosphotungstates
are more elucidative than the FT-IR data because this technique
is extremely sensitive to the Keggin POMs and the silica sup-
port does not show any band on the Raman region of these
compounds (Fig. 3). The FT-Raman spectra of phosphotungstates
compounds are very similar to the materials PW11@aptesSBA-15
and Co (PW )2@aptesSBA-15. The spectra of TBA salts of PW and
In spite of their different structures PW11 and Co (PW ) anions
4
9
were immobilized onto the functionalized silica large porous sup-
port aptesSBA-15. SBA-15 was chosen as support material because
the large surface area, easy surface modification with functional
groups, high stability and low toxicity. This support was firstly func-
tionalized with the organosilane (3-aminopropyl)triethoxysilane
4
9
11
(
to yield aptesSBA-15) by reaction of the hydroxyls groups of their
Co (PW ) and their corresponding composites show a strong band
near 990–950 cm attributed to ꢀs(W O) with terminal oxygen
[15,16,43]. Furthermore, it can be seen a slightly shift of this band
4 9 2
−
1
surface, in a post-synthesis step [35]. Then, phosphotungstate PW11
and Co (PW ) were immobilized on aptesSBA-15 surface. The
4
9 2
lacunary anions such as PW11 are highly nucleophilic and react eas-
ily with electrophilic groups [39]. The silicon atoms from the silanol
to lower wavenumbers when PW11 and Co (PW ) are immobi-
4 9 2
lized on aptesSBA-15. The Raman spectrum of PW presents other
1
1
−
1
(
Si OH) groups presented on the surface of silica materials present
band at 995 cm due to ꢀas(P O). The same is not clearly identi-
electrophilic properties. It was described in the literature that lacu-
nary anions can be immobilized on silica supports by covalent bond
fied for the sandwich Co (PW ) and also for the phosphotungstate
4
9 2
composites. It is well known that the introduction of metal cations
into the lacunary structure PW11 may decrease the splitting of the
last two bands [46].
established between the silanol groups and the POM via W
O Si
bonds [28]. The immobilization of the sandwich Co (PW ) may
4
9 2
occur by an electrostatic or covalent interaction between the cobalt
transition metal and the amine group of the support material.
Reflectance diffuse is also a helpful technique because the sup-
port aptesSBA-15 does not show any prominent band in the range
of 200–800 nm. The UV–vis/DRS spectra of lacunary PW11 and
the sandwich Co (PW ) and their respective composites mate-
[
39,40] The coordinating interaction between the water ligand and
Co in Co (PW ) is weak, consequently this coordinate site can eas-
4
9
2
4
9 2
ily be displaced by some ligands such as amine groups. The POM
easily bonds to the amine groups on surface of the support. The
amount of amine groups and phosphotungstates on the surface of
the material was established by elemental analysis (1.2 mmol of
rials PW11@aptesSBA-15 and Co (PW ) @aptesSBA-15 are drawn
in Fig. 4. All samples exhibit UV absorption maxima at ca. 250 and
300 nm which are attributed to oxygen-to-tungsten charge trans-
4 9 2
fer at terminal W O and W
O W bonds of phosphotungstate,
NH , 0.034 mmol of PW and 0.020 mmol of Co (PW ) in 1 g of
respectively [47]. The spectra of Co (PW ) (Fig. 4d) and the
2
11
4
9
2
4
9 2
material).
material Co (PW ) @aptesSBA-15 (Fig. 4e) can be compared to
4 9 2

9
8
S.S. Balula et al. / Catalysis Today 203 (2013) 95–102
(
e)
PW11@aptesSBA-15
(
d)
(
c)
PW11
-
40
-30
-20
-10
δ (ppm)
0
10
20
(
b)
a)
(
Fig. 5. 31P MAS NMR spectra of TBA salt PW11 (bottom) and PW11@aptesSBA-15
top).
(
1
200
900
600
300
Wavenumber (cm−1
)
aptesSBA-15 support was effective, and no degradation of the phos-
photungstate structures occurred.
Fig. 3. FT-Raman spectra of (a) aptesSBA-15, (b) PW11, (c) PW11@aptesSBA-15, (d)
Co4(PW9)2 and (e) Co4(PW9)2@aptesSBA-15, in the wavenumber region between
−1
3.2. Homogeneous catalysis
3
00 and 1900 cm
.
The oxidation of geraniol and R-(+)-limonene were carried out
in homogeneous phase using hydrogen peroxide, acetonitrile, and
catalytic amounts of TBA salts of the lacunary PW11, the mono-
evaluate possible alterations of CoII coordination environment of
the sandwich compound after the immobilization on silica. This
Co sandwich compound shows one d–d electronic transition band
with a maximum at ca. 580 nm with a shoulder at about 520 nm
and both are also observed when the compound is immobilized
II
substituted PW11M and the sandwich-type M (PW ) (M = Co and
4
9 2
III
Fe ). The TBA salts are essential to ensure their adequate solubility
in acetonitrile.
Geraniol (i) is an allylic alcohol that offer several possible places
of oxidative attach, namely at the two double bonds, at the allylic
(
Fig. 4e).
The integrity of lacunary PW11 compound was investigated
carbon centers and at the carbon of the CH OH group. The oxida-
2
before and after their immobilization on aptesSBA-15 by 31P MAS
NMR (Fig. 5) and in both spectra a single peak was observed. For TBA
salt of PW11 the peak appears at −10.8 ppm and for the compos-
ite PW11@aptesSBA-15 this is shifted to −11.3 ppm. This difference
tion of this substrate with H O2 at room temperature produces as
2
main product the 2,3-epoxygeraniol (ii) in the presence of all cata-
lysts, indicating the high chemo-selectivity nature of the reactions.
A small amount of diepoxide (iii) was found for all studied catalysts
after 1 h of reaction. However, the selectivity for the main product
was always higher than 70% (Scheme 1).
Table 1 shows the catalytic results for the different homoge-
neous catalysts tested for geraniol oxidation after 1.5 h of reaction.
The best catalytic performance was observed for the lacunary PW11
31
may be caused by the extreme sensitivity of P nucleus to its local
environment and by the interaction of PW11 with the support sur-
face. In conclusion, the comparison performed between the isolated
phosphotungstate and their composites using different techniques
indicates that the immobilization of PW11 and Co (PW ) on the
4
9 2
Fig. 4. Reflectance electronic spectra: (a) PW11, (b) PW11@aptesSBA-15, (c) PW11@aptesSBA-15-ac, (d) Co4(PW9)2, (e) Co4(PW9)2@aptesSBA-15 and (f) Co4(PW9)2@aptesSBA-
5-ac, where ac stands for after catalytic uses.
1

S.S. Balula et al. / Catalysis Today 203 (2013) 95–102
99
Table 1
Data from the oxidation of geraniol using different catalysts after 1.5 h of reaction.
a
−1
)
Selectivityb (%)
–
H2O2d efficiency (%)
Catalyst
Conversion (%)
TOFc (mol molcat−1
h
No catalyst
PW11
PW11Co
PW11Fe
Fe4(PW9)2
Co4(PW9)2
≈0
96
72
–
550
510
136
49
–
73
–
48
36
49
100
100
100
91
35 (89)e
30 (100)e
100
341
74
a
Reaction conditions: 1 mmol of geraniol, 4.5 mmol H2O2, 3 ␮mol phosphotungstate, 1.5 mL CH3CN, room temperature and light protection.
Based on the amount of consumed substrate; corresponding to 2,3-epoxigeraniol.
TOF = (moles of geraniol consumed)/[(moles of phosphotungstate) × 0.16 h].
b
c
d
e
H2O2 efficiency = [(moles of substrate oxidized/moles of H2O2 consumed) × 100].
Conversion data obtained after 3 h of reaction.
and for the sandwich Co (PW ) . In fact, the catalysts of iron
4
9 2
showed much lower conversion data than the cobalt catalysts at
least for short reaction times. Fig. 6 shows the kinetic profile of
the best catalysts and it is possible to conclude that the PW11 and
the Co (PW ) have similar kinetic profile. Both catalysts achieve
4
9 2
almost complete conversion after 1 h of reaction and at this stage
the selectivity remains 100% for 2,3-epoxide. The mono-substituted
PW11Co also achieve a complete conversion of geraniol oxidation
and 100% of selectivity for the 2,3-epoxide after 3 h of reaction. The
efficiency of H O2 usage was calculated for the oxidation of geran-
2
iol catalyzed by the different homogeneous catalysts (Table 1). The
highest H O efficiency is observed in the presence of the lacu-
2
2
II
III
nary PW11 what means that the presence of Co and Fe into the
phosphotungstate structure promote the decomposition of the oxi-
dant. The amount of no productive decomposition of H O is similar
2
2
Fig. 7. Conversion and selectivity for the oxidation of R-(+)-limonene catalyzed by
several phosphotungstates (after 3 h of reaction). The reaction conditions were:
3 ␮mol catalyst, 1 mmol R-(+)-limonene, 4.5 mmol H2O2, 1.5 mL CH3CN and tem-
for both transition metals and also no prominent difference was
observed for mono-substituted and sandwich compounds. In the
◦
perature 80 C.
presence of all catalyst, some amount of the added H O2 is still
2
present after 3 h of reaction. This fact could indicate that the used
amount of H O does not contribute for the lower performance of
some of the studied phosphotungstates.
2
2
for geraniol, in spite of the increase in temperature. Fig. 7 depicts
the catalytic performance for the studied phosphotungstates after
Few works has been published for the oxidation of geraniol cat-
alyzed by Keggin type polyoxometalate and most of them have been
executed using homogeneous systems [8,15,22]. Some of these
work was performed using metal mono-substituted borotungstates
as catalysts to oxidize selectively geraniol to 2,3-epoxygeraniol
under similar experimental conditions used in this work [8,15,22].
However, an improvement of the catalytic results was attained in
3
h of oxidation of R-(+)-limonene with H O . The best catalysts are
2 2
also the PW11 and the Co (PW ) . Their time courses are presented
4
9 4
in Fig. 8 and it can be noticed that the cobalt catalysts showed
similar kinetic profile. In the presence of PW11 catalyst the con-
version attains almost 100% after 3 h of reaction, and the reaction
catalyzed by this phosphotungstate achieves higher conversion for
less reaction time. The turnover frequency calculated after 10 min
of reaction for PW11 is much higher (869 mol mol
for the cobalt catalysts (around 70 mol molcat
ity for the different products of the R-(+)-limonene oxidation is
between 40% and 50% depending of the catalysts. Three products:
the presence of PW11 and Co (PW ) .
−1 −1
h
4
9
2
) than
). The selectiv-
cat
In the oxidation of R-(+)-limonene (iv, Scheme 1) the conversion
achieved for the same reaction time was lower than that obtained
−1 −1
h
1
00
100
8
0
0
0
0
0
8
0
0
0
0
0
6
4
2
6
4
2
PW1111
PW11
PW11Co
PW11 CC oo
11
PW11Fe
Co
4 9 2
(PW )
(
CP oW ( 9P ) W2 C )o 4
4 9 2
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Time (h)
0
0.5
1
1.5
2
2.5
3
Time (h)
Fig. 6. Kinetic profiles for the homogeneous oxidation of geraniol with H2O2 in
the presence of phosphotungstate catalysts. Reaction condition: 3 ␮mol catalyst,
1
mmol geraniol, 4.5 mmol H2O2, 1.5 mL CH3CN, room temperature and light pro-
Fig. 8. Kinetic profiles for the oxidation of R-(+)-limonene with H2O2 in the presence
of homogeneous lacunar and cobalt phosphotungstates catalysts.
tection.

1
00
S.S. Balula et al. / Catalysis Today 203 (2013) 95–102
Table 2
Performance of heterogeneous phosphotungstates catalysts for the oxidation of geraniol and R-(+) limonene with H2O2.
Catalyst
Time (h)
Geraniola
R-(+)-limoneneb
Conversion (%)
Conversion (%)
Selectivity (%)c
Selectivity (%)
0
aptesSBA-15
PW11@aptesSBA-15
24
4
0
50
96
33
77
0
100
100
100
100
0
57
90
8
d
42
d
2
4
4
4
54
Co4(PW9)2@aptesSBA-15
100e
e
2
15
95
a
Reaction conditions: 1 mmol of geraniol, 4.5 mmol of H2O2, 1.5 mL CH3CN, 87 mg of PW11@aptesSBA-15 or 150 mg of Co4(PW9)2@aptesSBA-15 (containing 3 ␮mol of
POM), room temperature, light protection.
b
Reaction conditions: 1 mmol of R-(+)-limonene, 4.5 mmol of H2O2, 1.5 mL CH3CN, 87 mg of PW11@aptesSBA-15 or 150 mg of Co4(PW9)2@aptesSBA-15 (containing 3 ␮mol
◦
of POM), 80 C.
c
Based on the amount of consumed substrate; corresponding to 2,3-epoxigeraniol.
Based on the amount of consumed substrate; corresponding to limonene-1,2-diol.
Selectivity for limonene-1,2-epoxide.
d
e
1
,2-epoxide (v), diepoxide (vi) and 1,2-diol (vii) were found. Ini-
hand, the active center(s) of the lacunary PW11 should be some or
all the five oxygen atoms from the lacuna region that are free to
coordinate or interact with solvents and other species. These com-
pounds are probably differently immobilized on the surface of
tially 1,2-epoxide is formed but after the first 30 min of reaction
the diol is already the main product in the presence of all catalysts.
The oxidation of this substrate with H O2 had been studied before
2
in a biphasic system (water-dichloroethane medium) catalyzed by
sandwich polyoxometalates [20,21]. These studies were performed
at low temperature but reasonable catalytic results were reached
only for long reaction times.
aptesSBA-15. The lacunary PW11 may be immobilized via W O Si
bonds as previously suggested in the literature [28] or by simple
electrostatic interaction with the surface. In opposite, the immobi-
lization of the sandwich Co (PW ) may occur by the interaction
4
9 2
The present study suggests that the lacunary PW11 is more effi-
cient homogeneous catalyst than the compounds with the insertion
of the cobalt center and the amine groups on the surface of the
material. The difference of catalytic activity observed for these
heterogeneous catalysts may be caused by different ways of immo-
bilization or different interaction between active centers and the
support surface.
II
III
of transition metal such as Co and Fe in its framework. This fact
could be related with the most efficient interaction between the
lacunary PW11 and H O than the mono-substituted and the sand-
2
2
wich POMs, probably due to the presence of the five oxygen atoms
in the lacunary region which are free to interact with other species.
The higher catalytic performance of the lacunary compound is an
advantage because this is the precursor of mono-substituted com-
pounds, what indicate that the pos-insertion of transition metals in
its structure is unnecessary to enhance high yields for the oxidation
Fig. 9 compares the kinetic profiles between the homogeneous
and heterogeneous catalysts for oxidation of R-(+)-limonene with
H O . The selectivity obtained for the oxidation of this substrate
2
2
in the presence of heterogeneous catalysts is still low because
identical amounts of limonene-1,2-diol (vii) and limonene-1,2-
epoxide (v) were found. In the presence of the heterogeneous
catalysts the formation of the diepoxide was practically insignif-
icant. The literature reports only an example of heterogeneous
catalytic oxidation of limonene by a silicotungstate immobilized
onto surface-modified SiO2 and in this case, 78% of yield to
limonene-8,9-epoxide was obtained after 24 h of reaction [48]. The
selectivity and the nature of the main product must be strongly
influenced by active center of the catalyst, as well as by the method-
ology of immobilization and the support.
of geraniol and R-(+)-limonene with H O .
2
2
3.3. Heterogeneous catalysis
The most active homogeneous catalysts for the oxida-
tion of geraniol and R-(+)-limonene with H O , PW11 and
2
2
Co (PW ) , were immobilized onto a functionalized silica sup-
4
9 2
port aptesSBA-15. Their performance as heterogeneous catalysts
was then investigated using the same experimental conditions
than in the homogeneous reactions. The catalytic performance
for PW11@aptesSBA-15 and Co (PW ) @SBA-15 is showed in
The solid catalysts can be easily separated by simple filtra-
tion followed by washing (with acetonitrile and n-hexane) and
4
9 2
Table 2.
1
00
80
As depict in Table 2, the oxidation reactions does not occurred
and no conversion was observed in the presence of the func-
tionalized support material (aptesSBA-15). The two heterogeneous
catalysts were active for the oxidation of geraniol and only the 2,3-
epoxigeraniol (ii) was formed. The PW11@aptesSBA-15 is slightly
PW11
6
4
0
0
more active catalysts than Co (PW ) @aptesSBA-15 since it was
4
9 2
PW11@aptesSBA-15
attained higher conversion after 4 h of reaction. Furthermore, after
4 h of reaction near complete conversion was achieved using
2
Co
4
(PW)
2
PW11@aptesSBA-15, however the same conversion was not real-
ized in the presence of Co (PW )2@aptesSBA-15. The difference of
Co
4
2
(PW) @aptesSBA-15
4
9
20
0
activity found between these phosphotungstates when immobi-
lized was not observed during their performance in homogeneous
systems. This difference of activity is still more pronounced for the
oxidation of R-(+)-limonene, where the Co (PW ) @aptesSBA-15
0
4
8
12
Time (h)
16
20
24
4
9 2
showed almost absence of activity, even after 24 h of reaction. The
active center of the sandwich Co (PW ) may be centered on the
4
9
2
2 2
Fig. 9. Kinetic profiles for the oxidation of R-(+)-limonene with H O in the presence
two cobalt metals with the terminal water ligand. On the other
of homogeneous and heterogeneous catalysts.

S.S. Balula et al. / Catalysis Today 203 (2013) 95–102
101
1
00
higher for Co (PW ) @aptesSBA-15 than for PW11@aptesSBA-15
4 9 2
system. These results may suggest that a partial leaching of the
active centers from the heterogeneous catalysts were lost for the
solution. The ICP analysis was performed to quantify the possi-
ble leaching of POM after three consecutive reaction cycles. Thus,
tungsten analysis was carried out for the PW11@aptesSBA-15 and
8
6
4
2
0
0
0
0
0
Co (PW ) @aptesSBA-15 recovered after three consecutive reac-
4
9 2
tion cycles, and no significant loss of tungsten content was found
for PW11@aptesSBA-15 (6.9 wt% of W before and 6.1 wt% after catal-
ysis). However, for Co4(PW9)2@aptesSBA-15 a loss of 27% of the
initial Co (PW ) was observed after the reaction cycles (4.9 wt% of
1
3
st Run
rd Run
2nd Run
4
9 2
leaching test
W before and 3.5 wt% after catalysis). To investigate the presence
of cobalt complex in solution after 24 h of heterogeneous catalytic
reaction, the solution from the reaction was collected and analyzed
0
4
8
12
Time (h)
16
20
24
II
by UV–vis spectroscopy, however the characteristic d–d band of Co
Fig. 10. Kinetic profiles of PW11@aptesSBA-15 for different recycling cycles (first,
second and third run) and for leaching analysis of geraniol oxidation reaction using
H2O2 as oxidant (leaching test).
was not observed.
The heterogeneous catalysts PW11@aptesSBA-15 and
Co (PW ) @aptesSBA-15 recovered after the reaction were
4
9 2
characterized by different techniques, in order to check the cat-
alysts stability after reaction cycles. The analysis of UV-vis/DRS
spectrum for PW11@aptesSBA-15 (Fig. 4c) presents the same
oxygen-to-tungsten charge transfer band as observed before
catalytic run. Also identical UV-vis/DRS spectrum was found for
dried at room temperature to be used in a fresh reaction under
identical experimental conditions. The catalysts were recycled in
order to test their stability and activity in consecutive reaction
cycles. Fig. 10 shows the reusability of the heterogeneous sys-
tem PW11@aptesSBA-15 for oxidation of geraniol during 24 h of
reaction. It can be seen that the activity of this catalyst is main-
tained during the consecutive cycles and the complete conversion
is achieved after 24 h of reaction. The same behavior was observed
for Co (PW ) @aptesSBA-15 where the conversion obtained after
Co (PW ) @aptesSBA-15 after its use in the catalysis reaction
4
9 2
(
Fig. 4f), showing the d–d transition band with a shoulder char-
II
acteristic of Co from sandwich compound. These results suggest
that the structure of PW11 and Co (PW ) remain intact after three
4
9 2
heterogeneous catalytic cycles. This stability was also confirmed
by FT-IR and by FT-Raman, where the characteristic bands of
4
9 2
2
4 h is still higher for the second (98%) and third cycles (95%) than
these phosphotungstates [␯as(W
O W) observed in FT-IR and
for the first cycle (77%). The selectivity for all the consecutives
cycles of both catalysts was always 100% for 2,3-epoxide of geran-
iol. To further investigate the activity of the heterogeneous catalyst
versus the possible homogeneous reaction with the leached of
active species from the support surface, the solid catalysts were
separated from the reaction mixture by carefully filtration after
␯s(W O) observed in FT-Raman] were also found after catalytic
reaction. The integrity of the support material was analyzed
by scanning electronic microscopy (SEM), and as observed in
Fig. 11 the morphology of the material is similar before and
after catalysis. Furthermore, the EDS analysis was performed to
determine the qualitative composition of the materials before
and after catalytic reaction. Tungsten was identified in the two
heterogeneous catalysts before and after catalytic uses. In the
2
h of reaction when the conversion was higher than 20% and the
reaction was continued with the remaining filtrate. This exper-
iment was performed for the oxidation of geraniol, and Fig. 10
also compares the kinetic profiles with and without filtration of
solid PW11@aptesSBA-15 after 2 h. It can be observed that after this
reaction time without solid catalyst the conversion decreases and
after 24 h the conversion was only 47% (leaching test) instead of
case of Co (PW ) @aptesSBA-15 cobalt was also identified, and
4
9 2
the ratio Co/W was compared before and after catalytic reac-
tion revealing similar values (wt% ratio W/Co = 19.7/1.12 before
catalytic uses and 20.0/1.26 after catalytic uses), what indicates
no decomposition of Co (PW ) structure. In conclusion, the
9
6% obtained in the presence of solid PW11@aptesSBA-15. Sim-
ilar experiment was performed for the same reaction catalyzed
by Co (PW )2@aptesSBA-15. In this case, the loss of activity by
4
9 2
results obtained from these different techniques indicate that
the support material is robust and stable after oxidative reac-
tions and also no decomposition of active phosphotungstates
occurred.
4
9
the solid catalysts filtration was only 15% what means that the
homogeneous contribution for the increases of conversion was
Fig. 11. Scanning electronic microscopy images of Co4(PW9)2@aptesSBA-15 before (left) and after (right) to be used as catalyst in the oxidation of geraniol with H2O2.

1
02
S.S. Balula et al. / Catalysis Today 203 (2013) 95–102
4
. Conclusions
[7] A.C. Estrada, M.M.Q. Simoes, I.C.M.S. Santos, M.G.P.M.S. Neves, J.A.S. Cavaleiro,
A.M.V. Cavaleiro, Mon. Chem. 141 (2010) 1223.
[
8] I.C.M.S. Santos, S.L.H. Rebelo, M.S.S. Balula, R.R.L. Martins, M. Pereira, M.M.Q.
Simoes, M.G.P.M.S. Neves, J.A.S. Cavaleiro, A.M.V. Cavaleiro, J. Mol. Catal. A:
Chem. 231 (2005) 35.
The reported work demonstrates that the phosphotungstates
ions, lacunary PW1 , mono-substituted PW11M and sandwich
1
M (PW ) (M is CoII or FeIII) are active homogeneous catalysts for
[9] M.M.Q. Simões, I.C.M.S. Santos, M.S.S. Balula, J.A.F. Gamelas, A.M.V. Cavaleiro,
M.G.P.M.S. Neves, J.A.S. Cavaleiro, Catal. Today 91–2 (2004) 211.
10] A.C. Estrada, M.M.Q. Simoes, I.C.M.S. Santos, M.G.P.M.S. Neves, A.M.S. Silva, J.A.S.
Cavaleiro, A.M.V. Cavaleiro, Appl. Catal. A: Gen. 366 (2009) 275.
[11] M.S.S. Balula, I.C.M.S. Santos, M.M.Q. Simões, M.G.P.M.S. Neves, J.A.S. Cavaleiro,
A.M.V. Cavaleiro, J. Mol. Catal. A: Chem. 222 (2004) 159.
12] I.C.M.S. Santos, M.M.Q. Simões, M.S.S. Balula, M.G.P.M.S. Neves, J.A.S. Cavaleiro,
A.M.V. Cavaleiro, Synlett (2008) 1623.
4
9
the oxidation of geraniol and R-(+)-limonene with H O . In partic-
2
2
[
ular, the lacunary and the cobalt catalysts demonstrated to be more
active than the iron compounds for the oxidation of both substrates.
II
Also, the lacunary and the mono-substituted PW11M (M = Co and
[
III
Fe ) showed to be more selective catalysts than the sandwich. The
2
,3-epoxide was the only product obtained from geraniol oxidation
[13] S.S. Balula, L. Cunha-Silva, I.C.M.S. Santos, F.A.A. Paz, J. Rocha, H.I.S. Nogueira,
A.M.V. Cavaleiro, in: A.T. Marques, A.F. Silva, A.P.M. Baptista, C. Sa, F. Alves, L.F.
Malheiros, M. Vieira (Eds.), Advanced Materials Forum Iv, vol. 587–588, Trans
Tech Publications Ltd., Stafa-Zurich, 2008, p. 538.
[14] A.C. Estrada, M.M.Q. Simoes, I.C.M.S. Santos, M.G.P.M.S. Neves, J.A.S. Cavaleiro,
A.M.V. Cavaleiro, ChemCatChem 3 (2011) 771.
catalyzed by PW11 and the mono-substituted PW11M. A complete
conversion of this substrate was achieved after 1.5 h of reaction
when catalyzed by PW11 and Co (PW ) . In the presence of the
last catalyst also the diepoxide was formed. The complete con-
version of R-(+)-limonene oxidation was also achieved after 3 h of
reaction when the lacunary PW11 was the catalyst. The selectivity
for limonene-1,2-diol was between 40% and 50% in the presence
of all catalysts and other two products were detected: limonene-
4
9 2
[
15] J.L.C. Sousa, I.C.M.S. Santos, M.M.Q. Simoes, J.A.S. Cavaleiro, H.I.S. Nogueira,
A.M.V. Cavaleiro, Catal. Commun. 12 (2011) 459.
[16] A.C. Estrada, I.C.M.S. Santos, M.M.Q. Simoes, M.G.P.M.S. Neves, J.A.S. Cavaleiro,
A.M.V. Cavaleiro, Appl. Catal. A: Gen. 392 (2011) 28.
[
17] A.C. Estrada, M.M.Q. Simoes, I.C.M.S. Santos, M.G.P.M.S. Neves, A.M.S. Silva, J.A.S.
Cavaleiro, A.M.V. Cavaleiro, Catal. Lett. 128 (2009) 281.
1
,2-epoxide and diepoxide. The lacunary PW11 and the sandwich
[18] M.M.Q. Simoes, C.M.M. Conceicao, J.A.F. Gamelas, P. Domingues, A.M.V. Cav-
aleiro, J.A.S. Cavaleiro, A.J.V. Ferrer-Correia, R.A.W. Johnstone, J. Mol. Catal. A:
Chem. 144 (1999) 461.
Co (PW ) showed the best homogeneous catalytic performance
and, consequently, were immobilized to prepared novel heteroge-
neous catalysts: PW11@aptesSBA-15 and Co (PW ) @aptesSBA-15
These phosphotungstates were immobilized for the first time
and their catalytic performance was analyzed. PW11@aptesSBA-
4
9 2
[
19] D. Mansuy, J.F. Bartoli, P. Battioni, D.K. Lyon, R.G. Finke, J. Am. Chem. Soc. 113
(1991) 7222.
4
9 2
[20] M. Bosing, A. Noh, I. Loose, B. Krebs, J. Am. Chem. Soc. 120 (1998) 7252.
[
21] S. Casuscelli, E. Herrero, N. Crivello, C. Perez, M.G. Egusquiza, C.I. Cabello, I.L.
Botto, Catal. Today 107–108 (2005) 230.
1
5 showed high catalytic activity for geraniol and R-(+)-limonene.
[22] I.C.M.S. Santos, M.M.Q. Simoes, M. Pereira, R.R.L. Martins, M.G.P.M.S. Neves,
J.A.S. Cavaleiro, A.M.V. Cavaleiro, J. Mol. Catal. A: Chem. 195 (2003) 253.
Both substrates were practically complete oxidized after 24 h of
reaction. For geraniol oxidation only the 2,3-epoxigeraniol was pro-
duced. However for R-(+)-limonene the 1,2-diol (yield 54%) was the
main product but 1,2-epoxide and diepoxide were also obtained.
The two heterogeneous catalysts were used in consecutive reac-
tion cycles without loss of activity, and their structural stability
was also investigated by their recovery after consecutive cycles and
characterization analysis by several spectroscopy and microscopy
techniques that indicate absence of major catalyst decomposition.
However, some homogeneous contribution in the heterogeneous
catalytic systems were detected by the leaching analysis what indi-
cate that the covalent attach of the phosphotungstate to the support
via amine group may not be stable and further study to improve
the covalent bond of polyoxometalates to silica materials need to
be performed.
[
23] L.F. Chen, K. Zhu, L.H. Bi, A. Suchopar, M. Reicke, G. Mathys, H. Jaensch, U. Kortz,
R.M. Richards, Inorg. Chem. 46 (2007) 8457.
24] N.V. Maksimchuk, M.S. Meigunov, Y.A. Chesalov, J. Mrowlec-Bialon, A.B. Jarzeb-
ski, O.A. Kholdeeva, J. Catal. 246 (2007) 241.
[
[25] R.F. Zhang, C. Yang, J. Mater. Chem. 18 (2008) 2691.
[26] J.L. Hu, K.X. Li, W. Li, F.Y. Ma, Y.H. Guo, Appl. Catal. A: Gen. 364 (2009) 211.
[27] Y. Zhou, R.L. Bao, B. Yue, M. Gu, S.P. Pei, H.Y. He, J. Mol. Catal. A: Chem. 270
(2007) 50.
[
[
[
[
28] X.D. Yu, L.L. Xu, X. Yang, Y.N. Guo, K.X. Li, J.L. Hu, W. Li, F.Y. Ma, Y.H. Guo, Appl.
Surf. Sci. 254 (2008) 4444.
29] A.P. Zhang, Y. Zhang, P.G. Wang, J. Li, Y. Lv, S.A. Gao, Res. Chem. Intermed. 37
(2011) 975.
30] D.K. Lyon, W.K. Miller, T. Novet, P.J. Domaille, E. Evitt, D.C. Johnson, R.G. Finke,
J. Am. Chem. Soc. 113 (1991) 7209.
31] M.S. Balula, J.A. Gamelas, H.M. Carapuca, A.M.V. Cavaleiro, W. Schlindwein, Eur.
J. Inorg. Chem. (2004) 619.
[32] I.C.M.S. Santos, J.A.F. Gamelas, M.S.S. Balula, M.M.Q. Simoes, M.G.P.M.S. Neves,
J.A.S. Cavaleiro, A.M.V. Cavaleiro, J. Mol. Catal. A: Chem. 262 (2007) 41.
[33] X. Zhang, Q. Chen, D.C. Duncan, R.J. Lachicotte, C.L. Hill, Inorg. Chem. 36 (1997)
381.
[34] M. Choi, F. Kleitz, D.N. Liu, H.Y. Lee, W.S. Ahn, R. Ryoo, J. Am. Chem. Soc. 127
2005) 1924.
[
4
(
Acknowledgments
35] C. Pereira, K. Biernacki, S.L.H. Rebelo, A.L. Magalhaes, A.P. Carvalho, J. Pires, C.
Freire, J. Mol. Catal. A: Chem. 312 (2009) 53.
The authors acknowledge the Funda c¸ ão para a Ciência e a Tec-
nologia (FCT, MEC, Portugal) for general funding, for the grant
no. Pest-C/EQB/LA0006/2011 (to Associated Laboratory REQUIMTE)
and for the R&D project PTDC/EQU-EQU/121677/2010.
[36] R.R.L. Martins, M.G.P.M.S. Neves, A.J.D. Silvestre, A.M.S. Silva, J.A.S. Cavaleiro, J.
Mol. Catal. A: Chem. 137 (1999) 41.
[
37] R.R.L. Martins, M. Neves, A.J.D. Silvestre, M.M.Q. Simoes, A.M.S. Silva, A.C. Tome,
J.A.S. Cavaleiro, P. Tagliatesta, C. Crestini, J. Mol. Catal. A: Chem. 172 (2001) 33.
[38] A.I. Vogel, A text-Book of Quantitative Inorganic Analysis Including Elementary
Instrumental Analysis, 3rd ed., Longmans, London, 1961.
[
[
[
39] Y.H. Guo, C.W. Hu, J. Mol. Catal. A: Chem. 262 (2007) 136.
40] J.A.F. Gamelas, D.V. Evtuguin, A.P. Esculcas, Transit. Met. Chem. 32 (2007) 1061.
41] X.J. Luo, C. Yang, Phys. Chem. Chem. Phys. 13 (2011) 7892.
References
[1] J.M. Derfe, M.M. Derfe, Encyclopedia of Chemical Technology Kirk-Orthmer,
[42] R.G. Finke, M.W. Droege, P.J. Domaille, Inorg. Chem. 26 (1987) 3886.
[43] C. Rocchiccioli-Deltcheff, M. Fournier, R. Franck, R. Thouvenot, Inorg. Chem. 22
(1983) 207.
[44] X.Y. Zhang, M.T. Pope, M.R. Chance, G.B. Jameson, Polyhedron 14 (1995) 1381.
[45] C.H. Chiang, H. Ishida, J.L. Koenig, J. Colloid Interface Sci. 74 (1980) 396.
[46] R.F. Klevtsova, E.N. Yurchenko, L.A. Glinskaya, L.I. Kuznetsova, L.G. Detusheva,
T.P. Lazarenko, J. Struct. Chem. 32 (1991) 538.
vol. 22, 3rd ed., Wiley, New York, 1978.
[
[
[
2] C.L. Hill, C.M. Prosser-McCartha, Coord. Chem. Rev. 143 (1995) 407.
3] R. Neumann, D. Juwiler, Tetrahedron 52 (1996) 8781.
4] M.T. Pope, A. Muller, Polyoxometalates From Platonic Solids to Anti-Retroviral
Activity, Kluwer, Dordrecht, 1994.
[5] C.L. Hill, Polyoxometalates in catalysis (thematic issue), J. Mol. Catal. A: Chem.
2
62 (2007) 1–2.
[47] K. Nomiya, Y. Sugie, K. Amimoto, M. Miwa, Polyhedron 6 (1987) 519.
[48] J. Kasai, Y. Nakagawa, S. Uchida, K. Yamaguchi, N. Mizuno, Chem. Eur. J. 12
(2006) 4176.
[
6] C.W. Jones, Applications of Hydrogen Peroxide and Derivatives, Royal Society
of Chemistry, Cambridge, 1999.