Silica-bound decatungstates as heterogeneous catalysts for H2O2 activation in selective sulfide oxidation

Article abstract of DOI:10.1016/j.jcat.2007.06.019

Alkylammonium decatungstates were covalently anchored to silica gel, furnishing efficient and robust heterogeneous catalysts able to activate H₂O₂ in the selective oxidation of sulfide to sulfoxide. The organic-inorganic hybrid materials were fully characterized, particularly by spectroscopic experiments. Primary propylammonium decatungstate was the most active catalyst. The use of low amounts of the heterogeneous catalyst (0.1 mol%) with a slight excess of 30% H₂O₂ (1.15 equiv.) in a nonchlorinated solvent (methanol) makes this oxidation reaction an environmentally benign chemical process. Leaching and recycling experiments revealed that the supported catalyst is not only highly efficient but also robust, because it can be used six times without loss of activity. Furthermore, the immobilized catalyst can tolerate different types of reaction solvents. Finally, this procedure was successfully applied to various sulfides, including allyl phenyl sulfide and dibenzyl sulfide.

Full text of DOI:10.1016/j.jcat.2007.06.019

Journal of Catalysis 250 (2007) 222–230
www.elsevier.com/locate/jcat
Silica-bound decatungstates as heterogeneous catalysts for H2O2 activation
in selective sulfide oxidation
Franca Bigi ∗, Andrea Corradini, Carla Quarantelli, Giovanni Sartori
Dipartimento di Chimica Organica e Industriale dell’Università, Parco Area delle Scienze 17/A, 43100 Parma, Italy
Received 13 April 2007; revised 16 June 2007; accepted 18 June 2007
Available online 30 July 2007
Abstract
Alkylammonium decatungstates were covalently anchored to silica gel, furnishing efficient and robust heterogeneous catalysts able to activate
H O in the selective oxidation of sulfide to sulfoxide. The organic–inorganic hybrid materials were fully characterized, particularly by spectro-
2
2
scopic experiments. Primary propylammonium decatungstate was the most active catalyst. The use of low amounts of the heterogeneous catalyst
(0.1 mol%) with a slight excess of 30% H O (1.15 equiv.) in a nonchlorinated solvent (methanol) makes this oxidation reaction an environmen-
2
2
tally benign chemical process. Leaching and recycling experiments revealed that the supported catalyst is not only highly efficient but also robust,
because it can be used six times without loss of activity. Furthermore, the immobilized catalyst can tolerate different types of reaction solvents.
Finally, this procedure was successfully applied to various sulfides, including allyl phenyl sulfide and dibenzyl sulfide.
© 2007 Elsevier Inc. All rights reserved.
Keywords: Heterogeneous catalysis; Decatungstate; Silica-supported decatungstate; Sulfide oxidation; Hydrogen peroxide activation
1. Introduction
only harmless water by reaction, is safe to store and use, and
is cheap and readily available.
Among the different metal catalysts, various types of tung-
sten-based catalytic systems have been reported to be used in
the activation of H2O2 [4]. Recently, Noyori and co-workers
described the activity of sodium tungstate combined with
phenylphosphonic acid and a quaternary ammonium hydrogen
sulfate as a phase-transfer catalyst [3a,5] in the oxidation of
alcohols, olefins, and sulfides.
A further step in the development of environmentally benign
chemical processes is the replacement of current homogeneous
oxidation procedures for the synthesis of fine chemicals by het-
erogeneous processes. Immobilized catalysts have been of great
interest due to several advantages, such as simplified product
workup, separation, isolation, and catalyst reuse [6]. Indeed,
chemical processes with little waste are expected from using
immobilized catalysts, because these catalysts can be easily re-
covered and reused.
Oxidation reactions are some of the most useful chemical
transformations despite the fact that they are among the most
problematic processes, due mainly to the production of large
amounts of pollutant materials [1]. At the international level,
significant effort has been devoted to substituting the traditional
stoichiometric inorganic oxidants such as Cr(VI) and Mn(VII)
with cleaner catalytic systems. The heavy metal oxidants give
toxic wastes, organic stoichiometric oxidants are usually very
expensive [2], and nitric acid unavoidably forms various nitro-
gen oxides.
Therefore, the need for cleaner and safer oxidation proce-
dures has prompted industrial and academic researchers to in-
vestigate the use of benign, easy-to-handle oxidants, such as
hydrogen peroxide and molecular oxygen. Indeed, the use of
hydrogen peroxide as an oxidant has attracted considerable at-
tention in recent years [3]. Aqueous hydrogen peroxide is an
ideal oxidant for liquid-phase reactions because it produces
Further developing our interest in heterogeneous catalysis
[7], in this work we present the preparation of alkyl ammonium
decatungstates chemically bound to silica support. We explore
the activity of these hybrid catalysts in H2O2 activation, exam-
ining the sulfide oxidation as a model reaction.
*
Corresponding author. Fax: +39 0521 905472.
E-mail address: franca.bigi@unipr.it (F. Bigi).
0021-9517/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2007.06.019

F. Bigi et al. / Journal of Catalysis 250 (2007) 222–230
223
Organosulfur compounds are versatile and useful intermedi-
ates in organic synthesis for the preparation of biologically ac-
tive products [8]. The sulfoxide group is present in well-known
drugs, the proton-pump inhibitors omeprazole and lansopra-
zole, as well as in the pesticide fipronil and various insecticides.
The selective oxidation of sulfides to sulfoxides is a problem
that remains the subject of intensive study [3e,9].
Polyoxometalates, high-valent early-transition metal–oxy-
gen anion clusters, compose a large class of inorganic com-
pounds with significant applications in several areas [10]. One
of the most extensively studied applications of polyoxometa-
lates has been in catalysis, where their use as both Brönsted
acid catalysts and oxidation catalysts began in the late 1970s
[11]. Different types of polyoxometalates have been used for
oxidative transformation with various oxygen donors. Oxida-
tion reactions with molecular oxygen and hydrogen peroxide as
“green” methods for organic synthesis have gained significant
impetus over the past decade [12].
(n-Bu4N)4W10O32 was prepared according to a literature pro-
cedure [19] and the adsorbed catalyst was prepared following
an “impregnation” procedure described previously [7a].
Na4W10O32 was prepared following a literature procedure
[13h], adding 260 mL of a boiling aqueous 1 M HCl solution to
a boiling solution containing Na2WO4·2H2O (44 g) in distilled
water (250 mL). The resulting solution was allowed to boil for
40 s, after which it was transferred to a 2-L beaker and rapidly
cooled to 0 ◦C in a liquid nitrogen/acetone bath under stirring.
Solid NaCl was added to saturation while the temperature was
maintained at 0 ◦C. A precipitate formed that was collected on a
fritted funnel; washed in a small amount of cold water, ethanol,
and diethyl ether; and transferred to a 250-mL beaker. (The use
of nonmetallic spatula is recommended to avoid the formation
of a blue color.) This precipitate was suspended in hot acetoni-
trile (130 mL); then the suspension was filtered, and the filtrate
was placed in a freezer overnight. Large pale-lime crystals of
sodium decatungstate were collected on a fritted funnel and
dried (9.4 g). From the mother liquor, it was possible to ob-
tain more crystals on concentration. The absorbance spectrum
in acetonitrile or in water comprised a well-defined maximum
at 324 or 323 nm, respectively.
The activity of the decatungstate anion W10O4− in promot-
ing photo-oxygenation of various substrates has32been studied
extensively [11b,13]. Recently, hexadecyl trimethyl ammonium
decatungstate was reported to catalyze hydrogen peroxide acti-
vation for the oxidation of alcohols [14], and a paper dealing
with lacunary polyoxotungstate for microwave-assisted H2O2
activation has been published recently [15].
Heterogenized alkylammonium decatungstates were previ-
ously reported in the literature as catalysts for photo-oxidation
reactions, but the immobilization of such catalysts was usually
performed by impregnation of the solid support [7,13c,16] or
by embedding in different polymeric membranes [13a,17a] or
inside the silica network [17b]. To enhance the activity and sta-
bility, it may be preferable to anchor the metal catalyst to the
support through a chemical bond. Thus, we designed an ammo-
nium cation to bound covalently to the solid support and then to
introduce the polyoxoanion via an exchange reaction [18]. The
anchoring of the metal catalyst through a chemical bond should
expand the range of possible reaction solvents to include those
solvents able to dissolve and to remove the ammonium de-
catungstate from the support. Furthermore, the chemical bond,
being stronger than electrostatic interactions, should increase
the robustness of the catalyst, making it reusable.
The starting materials for sulfide oxidation were hydro-
gen peroxide (30%, Carlo Erba), methyl phenyl sulfide (99%,
Aldrich), methyl-4-methoxyphenyl sulfide (97%, Aldrich),
methyl-4-bromophenyl sulfide (97%, Aldrich), dibenzyl sul-
fide (ꢀ95%, Fluka; it was recrystallized from diethyl ether),
diphenyl sulfide (98%, Aldrich), and allyl phenyl sulfide
(ꢀ96.5%, Fluka).
2.2. Catalyst preparation
The preparation of catalysts CAT I–IV involved two main
steps: (a) silica functionalization anchoring ammonium salts
to surface silanols and (b) anion exchange with sodium de-
catungstate (Scheme 1). Silica was activated by refluxing in
HCl conc. for 4 h, followed by washing until neutral with dis-
tilled water and then drying [20]. Alkylammonium heteroge-
nization was done by one of two procedures depending on the
type of ammonium cation.
Anchored primary, secondary, and tertiary ammonium salts
were prepared by refluxing activated silica (5 g) and the suit-
able (alkylaminopropyl)trialkoxysilane (10 mmol) in toluene
(30 mL) under stirring for 1 h. After distillation of a toluene
fraction containing ethanol, the refluxing was continued. After
1 h, this second procedure was repeated and refluxing contin-
ued for 0.5 h [21]. The cooled functionalized silica was filtered
off; washed with toluene, diethyl ether, and dichloromethane
(2 × 25 mL each); and then dried under high vacuum at 60 ◦C
for 3 h to give the surface-bound alkylamine groups with a
loading of 0.8–0.9 mmol/g. The resulting materials in dry
dichloromethane (5 g in 25 mL) were reacted with trifluo-
romethane sulfonic acid (two equivalents with respect to the
supported amino group) for 8 h at room temperature; filtered
off; washed successively with dichloromethane, ethanol, and
diethyl ether (2 × 25 mL each) [18a]; and then dried under
vacuum at 60 ◦C for 3 h, affording the corresponding alky-
In this paper we describe the preparation of various alkylam-
monium decatungstate catalysts immobilized on silica support.
2. Experimental
2.1. Materials
All materials purchased were used as such unless other-
wise stated. Starting materials for catalyst preparation: silica
gel KG60 (size, 0.040–0.063 mm; surface area, 480–540 m2/g;
pore volume, 0.74–0.84 cm3/g) for column chromatography
(Merck), (3-aminopropyl)triethoxysilane (99%, Aldrich), (3-
diethylaminopropyl)trimethoxysilane (ꢀ90%, Fluka), [3-meth-
ylaminopropyl]trimethoxysilane (ꢀ97%, Fluka), (3-bromo-
propyl)trimethoxysilane (ꢀ97%, Fluka), trifluoromethanesul-
fonic acid (ꢀ98%, Fluka), sodium tungstate dihydrate (Aldrich).

224
F. Bigi et al. / Journal of Catalysis 250 (2007) 222–230
Scheme 1.
lammonium salts bound to the solid surface (AS I–III in
Scheme 1, Ai). Elemental analyses revealed a sulfur loading
of 0.8–0.9 mmol/g.
and an aqueous solution of sodium decatungstate (4 mmol) in
distilled water at room temperature for 30 min (Scheme 1, B).
For example, supported propylammonium triflate AS I (5 g,
loading 0.80 mmol/g) was stirred with sodium decatungstate
(9.77 g, 4.0 mmol) in 30 mL of distilled water. The choice
of the reaction time is important to ensure the anion exchange
avoiding the degradation of sodium decatungstate in water as
evidenced by following the variation of the UV absorption band
at 323 nm. After stirring, the white solids were filtered off, care-
fully washed with 700 mL of distilled water, 50 mL of ethanol,
and 50 mL of diethyl ether. Then they were washed in contin-
uous with hot acetonitrile for 12 h using a Soxhlet apparatus.
After drying at 60 ◦C under high vacuum for 3 h, the catalysts
were completely characterized.
To prepare supported quaternary ammonium bromide, (3-
bromopropyl)trimethoxysilane (2.43 g, 10 mmol) was con-
densed with 5 g of silica silanols by refluxing in 30 mL of
toluene following the procedure described above. The bromo-
propylated silica (5 g, 0.5 mmol/g loading) was then treated
with triethylamine (1.01 g, 50 mmol) in 30 mL of refluxing
toluene for 24 h, affording the corresponding supported qua-
ternary ammonium bromide (AS IV in Scheme 1, Aii) [22].
After cooling, the solid was filtered on a Büchner funnel and
carefully washed with toluene (5 × 20 mL), then dried at 60 ◦C
under high vacuum. The amount of bromide ions present on
the functionalized silica after the reaction with triethylamine
was determined by titration [21] according to the method de-
scribed by Volhard, starting from 0.30 g of immobilized salt in
10 mL of ethanol, 10 mL of 0.1 N AgNO3 solution, and 5 mL
of HNO3 6 N were added, and the suspension was stirred in the
dark for 0.5 h at room temperature. Then the solid was filtered
off, and the excess AgNO3 was titrated with 0.1 N ammonium
thiocyanate, giving a bromide loading of 0.43 mmol/g.
The catalysts CAT I–IV were prepared by stirring a mixture
of the selected surface-bound alkylammonium salt (4 mmol)
2.3. Catalyst characterization
All functionalized materials were characterized with respect
to compositional, structural, and surface properties. The load-
ing of the organic groups was calculated by elemental analysis
performed with a Carlo Erba CHNS-O EA1108 elemental an-
alyzer. Metal elemental analyses were performed by ICP-AES
on an Ultima 2 Jobin Yvon HORIBA instrument, with testing
solutions prepared by dissolving about 5 mg of the catalyst in

F. Bigi et al. / Journal of Catalysis 250 (2007) 222–230
225
10 mL of 10% dilute NH4OH and further diluting by 1:10 with
distilled water. (It is important that these solutions be trans-
ferred quickly in PE vials when Na is determined, to avoid
contamination from glass.) EDS analyses of the catalysts, in a
windowless configuration, to determine the presence of sodium
and tungsten were performed in a Philips 515 scanning electron
microscope equipped with an EDAX Phoenix microanalyzer.
Thermogravimetric analysis and differential thermal analy-
sis (TGA–DTA) were carried out on a Perkin–Elmer TGA7
analysis system. N2 adsorption–desorption isotherms, obtained
at 77 K using a Micrometrics PulseChemiSorb 2705, were used
to determine specific surface areas, SABET. Before each mea-
surement, the samples were outgassed at 383 K for 1 h. FTIR
spectra of all of the catalysts (KBr pellets) were recorded on a
Nicolet FTIR Nexus spectrophotometer (resolution 4 cm−1) in
the range of 4000–400 cm−1. DRS-UV measurements were
performed using a Varian 2390 UV–vis spectrophotometer
equipped with an integrating sphere in the range of 220–
500 nm.
off [5]. The gas-chromatographic analyses were performed on
a TraceGC ThermoFinnigan instrument with a Supelco SPB-20
fused silica capillary column (30 m × 0.25 mm) with helium as
a carrier, adding 4-tert-butylphenol as an internal standard.
The hydrogen peroxide content in the reaction mixture was
measured following a standard iodometric titration method with
sodium thiosulfate [23]. The model reaction of methyl phenyl
sulfide and hydrogen peroxide was monitored between 10 and
120 min. Samples were obtained periodically, and the course of
the reaction was followed by gas chromatography using 4-tert-
butylphenol as the internal standard added to the samples.
The same methodology was followed in the synthetic ap-
plication to different sulfides using the best catalyst CAT I.
The products were purified by flash chromatography over sil-
ica gel column, using hexane–ethyl acetate mixtures as eluants.
All products gave melting points and spectral data consistent
with those reported.
3. Results and discussion
2.4. Reaction procedure for sulfide oxidation
3.1. Catalyst characterization
Typical oxidation of methyl phenyl sulfide with hydrogen
peroxide as model reaction was performed using a round-
bottomed flask. The selected catalyst (the amount of which was
evaluated on the basis of loading values for introducing the
specified amount of decatungstate) and 30% H2O2 (0.23 mL,
2.30 mmol) were added to a solution of methyl phenyl sulfide
(0.25 g, 0.23 mL, 2.0 mmol) in the specified solvent (10 mL).
The reaction was stirred at room temperature for 1.5 h. The
progress of the oxidation reaction was monitored by GC and
TLC. After 1.5 h, the mixture was filtered on Büchner fun-
nel. The solid catalyst was washed with 5 mL of methanol and
recovered. When the CH2Cl2/MeOH mixture was used as reac-
tion solvent, distilled water was added to the solution to obtain
phase separation, and further extraction was accomplished with
CH2Cl2 (2 × 10 mL). When the reaction solvent was MeOH,
the solution was rapidly evaluated by gas chromatography; oth-
erwise, Na2S2O3 was added to the solution to consume the
excess H2O2 (verified by an iodinated paper test) and filtered
Table 1 summarizes the BET surface areas. As expected, a
general decrease in surface area was found due to function-
alization, and a more pronounced effect was observed with
increasing of loading. Elemental analysis of the materials ob-
tained at each step of functionalization allowed us to evaluate
the loading of the supported organic moieties, which was found
to be 0.3–0.9 mmol/g. The higher values of loading observed
at the first step when propylamino groups were present could
be ascribed to base catalyzed condensation. The absence of sul-
fur atoms after the exchange process is noteworthy, indicating
the complete substitution of trifluoromethane sulfonate with de-
catungstate anions.
ICP-AES analyses of the catalysts allowed us to evaluate
the W loading. The W10 loading of catalysts CAT I–IV are
reported in Table 1. The simultaneous ICP-AES analyses of
W and Na demonstrated that practically all of the sodium
of Na4W10O32 was exchanged. The complete substitution of
sodium in supported decatungstate was confirmed by EDS mi-
Table 1
Surface area (SA
−
), A loading in precursor AS, N and decatungstate loading of catalysts CAT I–IV
BET
−
Catalyst
SA
(m /g)
A
load. in AS
N loading
(mmol/g)
W
loading
BET
10
2
(mmol/g)
(mmol/g)
CAT I
190
0.80
0.7
0.7
0.6
0.3
0.12
CAT II
CAT III
CAT IV
173
185
313
0.92
0.81
0.43
0.16
0.13
0.06

226
F. Bigi et al. / Journal of Catalysis 250 (2007) 222–230
Fig. 1. TGA/DTG analysis of CAT I.
Fig. 3. Diffuse reflectance UV–vis spectrum of CAT I.
Fig. 2. FT-IR spectra of (a) (n-Bu N) W
O ; (b) bare SiO ; (c) CAT I.
10 32 2
4
4
sorption at 324 nm (in addition to the band at 268 nm) in so-
lution assigned to oxygen-to-tungsten charge transfer (O → W
CT) transition. The UV–vis diffuse reflectance spectra of all
of the catalysts showed this typical band. Fig. 3 illustrates the
spectrum of CAT I.
croanalysis. The discrepancy between nitrogen and W10 load-
ings is not as high as it might appear, because one decatungstate
unit requires four ammonium cations. In addition, we verified
that silica treated with Na4W10O32 and washed according to
the exchange procedure did not present any amount of tung-
sten. This result indicates that the only species present on silica
are alkylammonium decatungstates bound to the support.
TGA/DTG of all of the catalysts revealed that the first step
of thermal decomposition, from room temperature to 170 ◦C,
corresponds to removal of the surface-adsorbed water, and that
the major weight loss occurs at high temperature, as expected
for chemisorbed materials, confirming that ammonium groups
are covalently bound on the surface of silica. Fig. 1 shows the
TGA/DTG results for CAT I.
FTIR spectra demonstrate that the structure of decatungstate
was still preserved on the silica surface. The characteristic fre-
quencies of W–O stretching modes [19,24a] were observed for
all the catalysts. Fig. 2 shows the spectra of the bare SiO2 (a),
(n-Bu4N)4W10O32 (b), and CAT I (c), in which the three bands
at 945, 893, and 803 cm−1 are readily distinguishable.
3.2. Catalytic activity
The catalytic activity of the heterogenized decatungstates
was tested in the oxidation of sulfides. The oxidation of phenyl
methyl sulfide with 30% H2O2 was selected as the model re-
action (Scheme 2). A CH2Cl2/CH3OH (1:1) mixture was first
chosen as reaction solvent because it gives a homogeneous
phase with the reagents, increasing the accessibility to the cata-
lyst, and avoids reagent/product adsorption on the solid support,
as was reported recently [18a].
We examined the catalytic activity of CAT I reacting methyl
phenyl sulfide (2 mmol) and 30% H2O2 in CH2Cl2/CH3OH
(1:1) (10 mL) at room temperature for 1.5 h. The amounts of
catalyst and hydrogen peroxide were varied to use the minimum
amount of each. The data, reported in Table 2, reveal that CAT I
is very active. Indeed, using a very low quantity (0.1 mol%) of
decatungstate with respect to sulfide, methyl phenyl sulfoxide
UV–vis spectroscopy is reported highly diagnostic for de-
catungstate polyanion [13h,24], showing a characteristic ab-

F. Bigi et al. / Journal of Catalysis 250 (2007) 222–230
227
Scheme 2.
Table 2
Methyl phenyl sulfoxide synthesis catalyzed by CAT I
CAT II, which contains a secondary ammonium group. We
can attribute this effect to the increasing steric hindrance of
the ions surrounding decatungstate, which make it less acces-
sible. Prompted by the high selectivity (ꢀ95%) observed for
all of the catalysts, we checked whether increasing the amount
of CAT IV (2 mol%) would produce more sulfoxide 3a in a
shorter time, due to the increased number of catalytic sites in
the reaction. After 45 min, the conversion was 95 with 96%
selectivity. The selectivity decreased during the time due to
overoxidation; after 90 min, it was 87% with complete con-
version (Table 3, entry 5). However, the TON values indicate
increasing the amount of catalyst decreased the efficiency, as
sometimes occurs [26]. Therefore, we found that all of the cat-
alysts prepared were able to promote the oxidation of sulfide 1
by hydrogen peroxide, increasing only the relative amount.
In an attempt to develop an environmentally friendly process,
we tried to avoid using chlorinated solvent. We succeeded in
accomplishing the reaction solely in methanol with good re-
sults, obtaining the solfoxide at 92% yield and 97% selectivity
(Table 4, entry 1). We also carried out the model reaction in iso-
propanol and acetonitrile as solvents and achieved the expected
sulfoxide 3a yields of 86 and 80%, respectively. These findings
demonstrate that methanol was the best reaction solvent. We
performed some experiments using 10% H2O2 and obtained
lower conversions (75% after 90 min and 83% after 120 min)
with high selectivity (96%). This negative role of water can
be attributed to competitive interaction with the metal and to
modification of the polarity of the catalyst surface. Indeed, the
increased surface hydrophobicity can decrease the reactivity of
the hydrophobic sulfide reagent.
The conversion and product distribution of the model reac-
tion were determined as functions of the reaction time (Fig. 4).
The results demonstrate that sulfoxide 3a was formed selec-
tively and the amount of sulfone 4a increased slightly increased
over time. It is interesting to note that sulfone 4a can be ob-
tained selectively (100%) in a quantitative yield reacting the
sulfide 3a with 3 equivalents of 30% H2O2 in the presence of
CAT I (1 mol%) for 4 h (Table 4, entry 2).
The possible hydrogen peroxide decomposition was exam-
ined both in the reaction mixture and in a blank experiment in
which hydrogen peroxide was stirred in the presence of CAT I
in methanol for 90 min. A hydrogen peroxide efficiency of 95–
98% was found, indicating that unproductive decomposition
was negligible.
a
b
Entry
CAT I
Ratio 1:2
3 yield
3 selectivity
TON
(mol%)
(%)
(%)
1
2
3
4
5
6
2
1:1.0
1:1.1
1:1.1
1:3.0
1:1.15
1:1.15
93
88
85
89
92
2
94
96
96
91
95
–
49
185
445
980
960
–
0.5
0.2
0.1
0,1
c
c
SiO
2
a
Reactions were performed reacting sulfide 2 mmol and 30% H O in
2
2
CH Cl /CH OH (1:1) (10 mL) for 90 min.
2
2
3
b
Turnover number calculated as products (mol)/catalyst (mol).
Reaction time 60 min, then the selectivity decreased.
c
Table 3
a
Effect of ammonium cation in CAT I–IV activity
b
Entry
CAT
3 yield (%)
TON
1
2
3
4
5
CAT I
92
49
32
30
960
515
333
315
50
CAT II
CAT III
CAT IV
CAT IV
c
87
a
Reactions were performed reacting sulfide 2 mmol, 30% H O 2.3 mmol,
2
2
catalyst 0.1 mol% (0.002 mmol of W ) in CH Cl /CH OH (1:1) (10 mL) for
10
2
2
3
90 min.
b
Turnover number calculated as products (mol)/catalyst (mol).
Catalyst 2 mol%.
c
was obtained at a 92% yield and 95% selectivity (Table 2, en-
try 5).
A slight excess of H2O2 (1.15 equiv.) was used when the
amount of catalyst was decreased. Using a greater amount of
H2O2 (3 equiv.) allowed us to obtain a high yield (89%) with
good, but decreased selectivity (91%) in a shorter time (1 h)
(Table 2, entry 4), after which the selectivity decreased due
to further oxidation to sulfone. Aiming to reduce the waste of
reagent, we used 1.15 equiv. of H2O2.
To investigate the possible influence of the solid support on
the reaction outcome, a blank reaction was carried out adding
unfunctionalized silica. Only traces (2%) of sulfoxide were ob-
tained (Table 2, entry 6).
The different catalysts CAT I–IV were tested to examine
the possible influence of the ammonium cation on the poly-
oxometalate activity. The amount of catalyst used in each ex-
periment was determined based on the loading value to intro-
duce the same supported decatungstate equivalents (0.1 mol%).
The data reported in Table 3 demonstrate a strong effect of
the ammonium cation. The catalyst containing primary ammo-
nium cation, CAT I, is the most active, giving phenyl methyl
sulfoxide at a very high yield and selectivity, followed by
To explore the general applicability of this reaction, we ex-
amined various sulfides. As reported in Table 4, variously sub-
stituted aryl methyl sulfides underwent smooth oxidation to
selectively afford the corresponding sulfoxides in high yields
(Table 4, entries 1–3). The selective oxidation of allyl phenyl
sulfide to the corresponding sulfoxide (Table 4, entry 4) is note-

228
F. Bigi et al. / Journal of Catalysis 250 (2007) 222–230
Table 4
Oxidation of sulfides 1 to sulfoxides 3 using 30% H O catalyzed by CAT I in
2
2
a
MeOH
b
Entry Sulfide
Product 3 yield
3 selectivity TON
(%)
(%)
c
1
2
3
3a
3b
3c
92
80
86
97
95
97
953
842
890
d
4
5
6
3d
3e
3f
83
94
56
96
94
95
865
995
588
e
e
Fig. 5. Catalyst recycling in methyl phenyl sulfide oxidation.
f
g
f
g
7
a
3f
90 (85)
90 (96)
995 (443)
ingly, under the described reaction conditions, even diaryl sul-
fide furnished the corresponding sulfoxides in very good yield
in a reaction carried out for 24 h, or for 6 h in the presence
of 0.2 mol% of catalyst (Table 4, entry 7). A possible mecha-
nism of this reaction involves the formation of peroxotungstate
species [15,25] and the subsequent nucleophilic attack of the
sulfur atom in the thioether on the peroxo species. Indeed, it
is known that thioethers are oxidized to sulfoxides by elec-
trophilic oxidants [27], which explains the minor reactivity of
aromatic thioethers, as found in other studies [28], due to delo-
calization of the electron density on the sulfur.
Reactions were performed reacting sulfide 2 mmol, 30% H O (1.15
equiv.), catalyst 0.1 mol% in CH OH (10 mL) for 90 min.
3
2
2
b
Turnover number calculated as products (mol)/catalyst (mol).
When the reaction was carried out using 30% H O (3 equiv.) and catalyst
1 mol% for 4 h the corresponding sulfone 4a was obtained in 99% yield and
c
2
2
100% selectivity.
d
No epoxidation product was detected.
e
f
1
Evaluated by H NMR.
Reaction time 24 h.
g
Reaction time 6 h using catalyst 0.2 mol%.
Considering our results all together, we can conclude that
heterogenized decatungstate CAT I is very efficient in the acti-
vation of hydrogen peroxide in sulfide oxidation. This catalyst
also shows high hydrogen peroxide efficiency for the reaction
studied.
3.3. Leaching test and recycling
To verify the heterogeneity of the catalytic process, we
performed the leaching test suggested by Lempers and Shel-
don [29]. The reaction was filtered after 15 min, and stirring
was continued for 75 min after the catalyst was removed. The
reaction yield did not change significantly (from 18 to 21%),
indicating that no leaching of the active catalytic species oc-
curred during the reaction. It is noteworthy that in contrast,
the (Bu4N)4W10O32 adsorbed on silica gave a positive leach-
ing test, with the product yield increasing from 20 to 60% after
filtration.
Finally, we carried out recycling experiments for the oxi-
dation of methyl phenyl sulfide 1a in methanol (Fig. 5). The
filtered catalyst was dried after washing and reused without
further activation. Interestingly, the recycled catalyst could be
reused for at least five reaction cycles with almost unchanged
results. For comparison, the adsorbed decatungstate demon-
strated a dramatic drop in yield (from 92 to 16%) during the
second cycle.
Fig. 4. Reaction profile of the oxidation of methyl phenyl sulfide 1a over CAT I.
worthy. Epoxidation of the double bond was not observed, and
the corresponding sulfoxide was obtained in good yield and se-
lectivity.
A surprising result was the selective oxidation of dibenzyl
sulfide (Table 4, entry 5). Indeed, it has been reported that the
oxidation of dialkyl sulfides leads to the formation of sulfones
along with sulfoxides with poor selectivity [3e,18a].
It is known that the formation of sulfoxides from diaryl sul-
fides is difficult to achieve using 30% H2O2 [5,18a]. Interest-

F. Bigi et al. / Journal of Catalysis 250 (2007) 222–230
229
4. Conclusion
[6] (a) D.C. Bailey, S.H. Langer, Chem. Rev. 81 (1981) 109;
(b) A. Akelah, D.C. Scherrington, Chem. Rev. 81 (1981) 557;
(c) J.M. Fréchet, Tetrahedron 37 (1981) 663;
Heterogenized ammonium decatungstates bound to silica
support were developed by an exchange process on anchored
alkylammonium cations and used as catalysts to activate hydro-
gen peroxide in sulfide oxidation. Spectroscopic studies demon-
strated that the decatungstate structure was preserved in the
catalysts. The best catalyst, CAT I, was very efficient, giving
methyl phenyl sulfoxide in high yield (92%) and selectivity
(95%) under mild heterogeneous conditions. Our goal of ob-
taining a robust catalyst linking the polyoxometalate to the solid
support by chemical bonds rather than by simple electrostatic
interactions was achieved. Indeed, Sheldon’s test demonstrated
that no leaching of the active species occurred and the catalyst
could be reused at least five times. Furthermore, different reac-
tion solvents could be used without removing the decatungstate
from the support. In addition, there is the advantage of easy
separation and recovery of the heterogeneous catalyst.
(d) S. Kobayashi, R. Akiyama, Chem. Commun. (2003) 449.
[7] (a) A. Maldotti, A. Molinari, G. Varani, M. Lenarda, L. Storaro, F. Bigi,
R. Maggi, A. Mazzacani, G. Sartori, J. Catal. 209 (2002) 210;
(b) G. Sartori, F. Bigi, R. Maggi, R. Sartorio, D.J. Macquarrie, M. Lenarda,
L. Storaro, S. Coluccia, G. Martra, J. Catal. 222 (2004) 410;
(c) F. Bigi, L. Moroni, R. Maggi, G. Sartori, Chem. Commun. (2002) 716;
(d) G. Sartori, F. Bigi, R. Maggi, A. Mazzacani, G. Oppici, Eur. J. Org.
Chem. (2001) 2513;
(e) R. Ballini, F. Bigi, E. Gogni, R. Maggi, G. Sartori, J. Catal. 191 (2000)
348.
[8] (a) P. Metzner, A. Thuillier, Sulfur Reagents in Organic Synthesis, Acad-
emic Press, London, 1994;
(b) J. Drabowicz, P. Kielbasinski, M. Mikolajczyk, in: S. Patai, Z. Rap-
poport, C. Stirling (Eds.), The Chemistry of Sulfones and Sulfoxides,
Wiley, Chichester, UK, 1988, p. 233;
(c) I. Fernandez, N. Khiar, Chem. Rev. 103 (2003) 3651.
[9] (a) A.A. Lindén, L. Krüger, J.-E. Bäckvall, J. Org. Chem. 68 (2003) 5890,
and references herein.
[10] A special issue of Chemical Reviews is devoted to polyoxometalates, see:
in: C.L. Hill (Ed.), Chem. Rev. 98 (1998) 1.
The oxidation procedure using CAT I is of general applica-
bility to various sulfides with very good yields and selectivities
that are similar or higher than those reported in the literature.
Moreover, this catalyst shows high hydrogen peroxide effi-
ciency for the reaction studied.
[11] (a) I.V. Kozhevnikov, Catalysis by Polyoxometalates, Wiley, Chichester,
England, 2002;
(b) C.L. Hill, C.M. Prosser-McCartha, Coord. Chem. Rev. 143 (1995) 407;
(c) T. Okuhara, N. Mizuno, M. Misono, Adv. Catal. 41 (1996) 113.
[12] (a) M.V. Vasylyev, R. Neumann, J. Am. Chem. Soc. 126 (2004) 884;
(b) N.V. Maksimchuk, M.S. Melgunov, Yu.A. Chesalov, J. Mrowiec-
Białon´, A.B. Jarze˛bski, O.A. Kholdeeva, J. Catal. 246 (2007) 241.
The use of a low amount of heterogeneous catalyst (0.1
mol%) and a slight excess of H2O2 (1.15 equivalent) in a
nonchlorinated solvent, along with the absence of additives
(such as soluble phase transfer catalyst) or cocatalyst (such as
C6H5–PO3H2) makes this oxidation reaction an environmen-
tally benign chemical process.
[13] (a) M. Carraro, M. Gardan, G. Scorrano, E. Orioli, E. Fontananova, M.
Bonchio, Chem. Commun. (2006) 4533, and references herein;
(b) D. Dondi, M. Fagnoni, A. Molinari, A. Maldotti, A. Albini, Chem. Eur.
J. 10 (2004) 142;
(c) A. Molinari, R. Amadelli, A. Mazzacani, G. Sartori, A. Maldotti, Lang-
muir 18 (2002) 5400;
(d) I.N. Lykakis, C. Tanielian, R. Seghrouchni, M. Orfanopoulos, J. Mol.
Catal. A Chem. 262 (2007) 176;
Acknowledgments
This work was supported by the Ministero dell’Università e
della Ricerca (MiUR), Italy, and the University of Parma (Na-
tional Project “Attivazione ossidativa catalitica e fotocatalitica
per la sintesi organica”). The authors thank Professor R. Capel-
letti and Mr. C. Mora for help with the DRS-UV measurements,
Dr. D. Calestani for EDX microanalyses, and Dr. G. Gnappi for
TGA–DTA measurements. They also thank the Centro Inter-
dipartimentale Misure (CIM) for the use of their NMR instru-
ments.
(e) C. Tanielian, F. Cougnon, R. Seghrouchni, J. Mol. Catal. A Chem. 262
(2007) 164;
(f) A.J. Bridgeman, G. Cavigliasso, J. Phys. Chem. A. 106 (2002) 6114;
(g) C. Tanielian, R. Seghrouchni, C. Schweitzer, J. Phys. Chem. A 107
(2003) 1102;
(h) D.C. Duncan, T.L. Netzel, C.L. Hill, Inorg. Chem. 34 (1995) 4640.
[14] M.-L. Guo, Green Chem. 6 (2004) 271.
[15] M. Carraro, L. Sandei, A. Sartorel, G. Scorrano, M. Bonchio, Org. Lett. 8
(2006) 3671.
[16] A. Molinari, R. Amadelli, V. Carassiti, A. Maldotti, Eur. J. Inorg. Chem.
(2000) 91.
[17] (a) M. Bonchio, M. Carraio, G. Scorrano, E. Fontananova, E. Orioli, Adv.
Synth. Catal. 345 (2003) 1119;
References
(b) Y. Guo, C. Hu, X. Wang, Y. Wang, E. Wang, Y. Zou, H. Ding, S. Feng,
Chem. Mater. 13 (2001) 4058.
[1] R.A. Sheldon, J.K. Kochi, Metal-Catalyzed Oxidations of Organic Com-
pounds, Academic Press, New York, 1981.
[2] B.M. Trost, I. Fleming (Eds.), Comprehensive Organic Synthesis, vol. 7,
first ed., Pergamon, Oxford, 1991.
[18] (a) B. Karimi, M. Ghoreishi-Nezhad, J.H. Clark, Org. Lett. 7 (2005) 625;
(b) A. Molinari, G. Varani, E. Polo, S. Vaccari, A. Maldotti, J. Mol. Catal.
A Chem. 262 (2007) 156.
[3] (a) R. Noyori, M. Aoki, K. Sato, Chem. Commun. (2003) 1977;
(b) Z.W. Xi, N. Zhou, Y. Sun, K.L. Li, Science 292 (2001) 1139;
(c) K. Sato, M. Aokil, R. Noyori, Science 281 (1998) 1646;
(d) Y.Q. Deng, Z.F. Ma, K. Wang, J. Chen, Green Chem. 1 (1999) 275;
(e) V. Cimpeanu, V.I. Parvulescu, P. Amorós, D. Beltrán, J.M. Thompson,
C. Hardacre, Chem. Eur. J. 10 (2004) 4640.
[4] (a) Y. Ishii, H. Tanaka, Y. Nisiyama, Chem. Lett. (1994) 1;
(b) R. Neumann, D. Juwiler, Tetrahedron 52 (1996) 8781;
(c) F.M. Collins, A.R. Lucy, C. Sharp, J. Mol. Catal. 117 (1997) 397.
[5] K. Sato, M. Yodo, M. Aoki, X.Q. Zheng, R. Noyori, Tetrahedron 57 (2001)
2469.
[19] A. Chemseddine, C. Sanchez, J. Livage, J.P. Launag, M. Fournier, Inorg.
Chem. 23 (1984) 2609.
[20] G. Desaedeleer, C. Ronneau, D. Apers, Anal. Chem. 48 (1976) 570.
[21] P. Tundo, P. Venturello, J. Am. Chem. Soc. 101 (1979) 6606.
[22] T.M. Jyothi, M.L. Kaliya, M. Herskowitz, M.V. Landau, Chem. Commun.
(2001) 992.
[23] A.I. Vogel, Vogel’s Textbook of Quantitative Chemical Analysis, fifth ed.,
Longman, Harlow, 1989, p. 394.
[24] (a) F. Chauveau, M. Boyer, B. Le Meur, C.R. Acad. Sci. Ser. C 268 (1969)
479;

230
F. Bigi et al. / Journal of Catalysis 250 (2007) 222–230
(b) S.C. Termes, M.T. Pope, Inorg. Chem. 2 (1978) 500.
(b) M. Bonchio, S. Campestrini, V. Conte, F. Di Furia, S. Moro, Tetrahe-
dron 51 (1995) 12363.
[25] B.F. Sels, D.E. De Vos, P.A. Jacobs, J. Am. Chem. Soc. 123 (2001)
[28] (a) V. Hulea, P. Moreau, F. Di Renzo, J. Mol. Catal. A Chem. 111 (1996)
325;
(b) V. Huleau, P. Moreau, J. Mol. Catal. A Chem. 113 (1996) 499.
[29] H.E.B. Lempers, R.A. Sheldon, J. Catal. 175 (1988) 62.
8350.
[26] D.-S. Tong, J.Y. Wang, H.-Y. Niu, G.-Y. Wang, J. Mol. Catal. A Chem. 268
(2007) 120.
[27] (a) F. Di Furia, G. Modena, Pure Appl. Chem. 54 (1982) 1853;