A new halide-free efficient reaction-controlled phase-transfer catalyst based on silicotungstate of [(C18H37)2(CH 3)2N]3[SiO4H(WO5) 3] for olefin epoxidation, oxidation of sulfides and alcohols with hydrogen peroxide

Article abstract of DOI:10.1039/c4ra04036h

A new reaction-controlled phase-transfer catalyst based on silicotungstate of [(C₁₈H₃₇)₂(CH₃) ₂N]₃[SiO₄H(WO₅)₃] for oxidation of hydrocarbons is developed. The catalyst is a new heteropoly compound with silicon as heteroatom, which is different to the previously reported reaction-controlled phase transfer catalysts that were composed of quaternary ammonium heteropolyoxotungstates of [π-C₅H ₅N(CH₂)₁₅CH₃]₃[PW ₄O₁₆] and [π-C₅H₅N(CH ₂)₁₅CH₃]₃[PW₄O ₃₂] with phosphorus as heteroatom. The oxidation of various alkenes (such as linear terminal olefins, internal olefins, cyclic olefins and unactivated alkenes) to epoxides, sulfides to sulfoxides and sulfones, alcohols to carbonyl compounds, are successfully catalyzed by this recyclable and environmentally benign catalyst using H₂O₂ as oxidant and ethyl acetate as solvent. This catalyst is not only capable of catalyzing homogeneous oxidation of organic substrates with unique reaction-controlled phase-transfer character, but also avoids the use of toxic solvents. The catalyst could be easily recovered and reused after reaction, and the epoxidation of cyclohexene was performed twenty times without obvious loss in activity. The fresh catalyst and the used one were characterized by ICP, IR, UV-vis, ²⁹Si MAS NMR and ¹⁸³W NMR in detail. the Partner Organisations 2014.

Full text of DOI:10.1039/c4ra04036h

RSC Advances
View Article Online
View Journal | View Issue
PAPER
A new halide-free efficient reaction-controlled
phase-transfer catalyst based on silicotungstate of
Cite this: RSC Adv., 2014, 4, 32054
[
(C H ) (CH ) N] [SiO H(WO ) ] for olefin
1
8 37 2
3 2
3
4
5 3
epoxidation, oxidation of sulfides and alcohols with
hydrogen peroxide†
a
b
a
a
a
Baochun Ma,* Wei Zhao, Fuming Zhang, Yingshuai Zhang, Songyun Wu
a
and Yong Ding
A new reaction-controlled phase-transfer catalyst based on silicotungstate of [(C18
37 2 3 2 3
H ) (CH ) N] -
[
SiO H(WO ] for oxidation of hydrocarbons is developed. The catalyst is a new heteropoly compound
4
5 3
)
with silicon as heteroatom, which is different to the previously reported reaction-controlled phase
transfer catalysts that were composed of quaternary ammonium heteropolyoxotungstates of
5 5 2 3 3 4 5 5 2 3 3 4
[p-C H N(CH )15CH ] [PW O16] and [p-C H N(CH )15CH ] [PW O32] with phosphorus as heteroatom.
The oxidation of various alkenes (such as linear terminal olefins, internal olefins, cyclic olefins and
unactivated alkenes) to epoxides, sulfides to sulfoxides and sulfones, alcohols to carbonyl compounds,
2 2
are successfully catalyzed by this recyclable and environmentally benign catalyst using H O as oxidant
and ethyl acetate as solvent. This catalyst is not only capable of catalyzing homogeneous oxidation of
organic substrates with unique reaction-controlled phase-transfer character, but also avoids the use of
toxic solvents. The catalyst could be easily recovered and reused after reaction, and the epoxidation of
cyclohexene was performed twenty times without obvious loss in activity. The fresh catalyst and the
Received 12th May 2014
Accepted 27th June 2014
DOI: 10.1039/c4ra04036h
29
183
www.rsc.org/advances
used one were characterized by ICP, IR, UV-vis, Si MAS NMR and
W NMR in detail.
18–21
were used as solvents in their systems. Later, Mizuno
developed a series of silicotungstate polyoxometalates as effi-
N) [g-
40]. Both the
alkenes is among the most important reactions in organic activities and selectivities of these homogeneous catalytic
Introduction
Epoxides are highly useful intermediates for the manufacture of cient catalysts for alkene epoxidation, such as (Bu
a range of important commercial products, and epoxidation of SiW10 O) ] and (Bu N) [g-1,2-H SiV
4
4
O34(H
2
2
4
4
2
2 10
W O
1–4
synthesis. A catalytic epoxidation system that uses hydrogen epoxidation of olens with H O were excellent. However, there
2 2
peroxide as an oxidant is very attractive from the viewpoint of was a common problem for the difficulty in separation and
industrial production and synthetic organic chemistry since reuse of catalysts in these homogeneous catalytic systems.
aqueous hydrogen peroxide is cheap, environmentally clean
and easy to handle.
The searching for more efficient catalytic systems that
The chemistry of polyoxometalates combine the advantages of both homogeneous and heteroge-
5
–8
(
POMs) has drawn wide attention in the last two decades as neous catalysis is one of the most exciting challenges of modern
8–27
22–24
effective catalysts for epoxidation of alkenes.
The groups of chemistry. 2001, Xi
reported a reaction-controlled phase-
N(CH [PW based on
15,16
17,27
Venturello
and Ishii
independently developed highly transfer catalyst [p-C
H
5 5
)15CH ]
2 3 3
O ]
4 16
effective tungstophosphate-based catalyst systems for alkene quaternary ammonium heteropolyoxotungstates, which exhibited
epoxidation with hydrogen peroxide. However, toxic and carci- high conversion and selectivity for catalytic epoxidation of olens.
nogenic chlorocarbons (chloroform and 1,2-dichloroethane) Aer reaction, the catalyst can be recovered and reused just like a
8,26
heterogeneous catalyst by ltration. 2008, our group reported
another efficient reaction-controlled phase-transfer catalytic
State Key Laboratory of Applied Organic Chemistry and College of Chemistry and system based on quaternary ammonium heteropolyoxotungstate
a
Chemical Engineering, Lanzhou University, Lanzhou 730000, China. E-mail:
5 5 2 3 3 4
of [p-C H N(CH )15CH ] [PW O32]. The above reaction-controlled
mabaochun@lzu.edu.cn
b
phase-transfer catalysts were both based on polyoxometalate of
which the heteroatom was phosphorus. Does tungsten-based
polyoxometalate catalyst with silicon as the heteroatom has the
reaction-controlled phase-transfer catalytic character?
College of Chemistry and Pharmaceutical Engineering, Nanyang Normal University,
Nanyang 473061, China
†
Electronic supplementary information (ESI) available: Experimental details and
spectrum characterizations of catalysis reaction. See DOI: 10.1039/c4ra04036h
32054 | RSC Adv., 2014, 4, 32054–32062
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
In an attempt to interpret and address this question, we Varian mercury 300 MHz spectrometer with TMS as an internal
investigated the epoxidation of 1-octene with H O catalyzed by standard and CDCl as solvent unless otherwise noted.
2
2
3
a series of silicotungstates in ethyl acetate at 333 K as model
tests. Aer screening many different silicotungstates, we found Preparation of the catalyst
that a dioctadecyl dimethyl ammonium (DDA) salt of silico-
tungstate, [(C18H ) (CH ) N] [SiO H(WO ) ] (I) showed good
37 2 3 2 3 4 5 3
K [b -SiW O ]$14H O and K [g-SiW O ]$12H O were
8
2
11 39
2
8
10 36
2
synthesized according to ref. 28. [(C H ) (CH ) N] [SiO H-
1
8
37 2
3 2
3
4
catalytic activity over the epoxidation of various alkenes with the
reaction-controlled phase-transfer feature that could be recov-
ered and reused just like a heterogeneous catalyst aer reaction.
The epoxidation was conducted in ethyl acetate, and many
kinds of alkenes could be converted to the corresponding
epoxides in high yields. In an aqueous H O /ethyl acetate
biphasic system, catalyst I itself was not soluble in the ethyl
acetate phase, but it dissolved slowly with the increasing of
reaction temperature and with the addition of hydrogen
peroxide, which subsequently led to homogeneous catalytic
system for epoxidation of olens. When the reaction nished
with the decrease of temperature, catalyst I precipitated as a
white solid from the solvent gradually. This phenomenon
manifested a solid–liquid–solid phase transfer of catalyst I
controlled by the reaction conditions in ethyl acetate. Various
suldes and alcohols were also transformed into their corre-
sponding oxidative forms over this reaction-controlled phase-
transfer catalyst.
(
WO ) ]:K [g-SiW O ]$12H O (2.5 g, 0.85 mmol) was dissolved
5 3 8 10 36 2
ꢀ1
in 40 ml 30% H
2
O
2
(H
2
O
2
/K
8
[g-SiW10
O
36]$12H
2
O ¼ 450 mol 1
ꢀ1
mol ) and the solution was stirred for 4 h at room temperature.
Then 48 ml 30% H
1
2
O
2
(H
2
O
2
/K
8
[g-SiW10
O
36]$12H
2
O ¼ 550 mol
ꢀ1
ꢀ1
mol ) was added to the above solution. Subsequently, to
2
2
this solution, 1 g (1.7 mmol) of dioctadecyl dimethyl ammo-
nium chloride (DDAC) dissolved in 25 ml of tert-butanol was
added. The mixture was stirred vigorously for 4 h at 318 K. A
white occule was ltered off and then washed with an excess
amount of warm water (ꢁ318 K, about 100 ml) and dried in
37 2 3 2 3 4 5 3
vacuum. [(C18H ) (CH ) N] [SiO H(WO ) ] was obtained as a
white powder. Yield: 1.1 g (75%, based on the quaternary
ammonium salt charged). Anal. calcd for [(C18 (CH N]
SiO H(WO ]: C, 53.98; H, 9.51; N, 1.66; Si, 1.10; W, 21.76%;
H
37
)
2
3
)
2
3
-
[
4
5 3
)
Found: C, 52.42; H, 9.51; N, 1.75; Si, 1.09; W, 21.23%. IR spec-
trum (KBr, cm ): 2921, 1717, 1467, 1088, 1055, 1000, 967, 942,
8
ꢀ1
46, 722, 634, 574, 511, 473, 448. UV-vis spectrum (in CDCl )
3
29
showed shoulder bands at 243 nm. Si MAS NMR (79.5 MHz):
1
83
ꢀ
79.2 ppm. W NMR (16.6 MHz, CDCl
Active oxygen was 1.7% in our catalyst [(C18
SiO H(WO ]. Active oxygen was analyzed by the improved
iodometric method proposed by Venturello.
3
, 300 K): ꢀ316 ppm.
H
37
)
2
(CH N] -
3
)
2
3
Experimental
Reagents
[
4
5 3
)
16
All chemicals were analytical grade, commercially available and
Other polyoxometalates preparation methods see the ESI.†
used without further purication unless otherwise stated.
Catalytic reaction
Characterization techniques
The catalytic reactions were performed in a 10 ml round-
bottomed ask,
condenser. The epoxidation was carried out as follows: catalyst
10 mmol), solvent (2 ml), substrate (2.5 mmol), and H O (30%
a magnetic stirring bar, and a reux
Infrared spectra of heteropolyoxotungstates were recorded on a
Nicolet NEXUS 670 FT-IR spectrometer. The W NMR spectra
1
83
(
2
2
were recorded in 5 mm o.d. tubes enclosed a capillary lled with
1
83
aq., 0.5 mmol) were charged in the reaction ask. The reaction
was carried out at 333 K and detected by TLC accompanied with
GC. The yields of products were calculated from the peak areas
using an internal standard method by GC. The products were
identied by GC/MS (Finnigan Trace DSQ). Aer reaction
(reaction temperature decreased to room temperature), the
solid catalyst precipitated from the solvent and was separated
by centrifugation. The recovered catalyst can be used for the
next epoxidation aer the treat with wash and dry.
The reactor for epoxidation of propylene was a 250 ml Teon
coated steel autoclave and the reaction was carried out as
2 2
follows: catalyst (0.185 mmol), ethyl acetate (100 ml), H O (30%
aq., 20 mmol) and propylene (4 atm) were charged in the
autoclave. The reaction was maintained at 333 K for 2 hours.
Resulting products were applied to GC analysis.
C D . The W chemical shis were measured with respect to
6
6
an external 2 M Na WO solution in alkaline D O by using
2
4
2
dodecaphosphotungstic
salt
of
[p-C H N(CH ) CH ] -
5
5
2 11
3 3
[
(
PW12O40] in dichloromethane solution as a secondary standard
2
9
d ꢀ85.0 ppm). Si MAS NMR spectra were recorded at 9.4 T on
29
a Bruker Avance-400 wide bore spectrometer. The Si MAS
NMR spectra of solid catalyst with cross-polarization were per-
formed at 79.5 MHz with BB (broad band) MAS probe head
using 4 mm ZrO rotors and 2 ms constant time and 3 s repe-
2
tition time and 11 600 scans, with samples spun at 10 kHz and
referenced to SiMe . Chemical elemental analysis of the cata-
4
lysts was done as follows: Si, W content was analyzed by PE 4300
DV ICP-AES spectrometer. C, H, and N contents were measured
on a German Elementar Vario EL spectrometer. GC analysis was
performed on Shimadzu GC-9AM with a ame ionization
detector equipped with SE-54 capillary (internal diameter ¼ 0.25
mm, length ¼ 30 m). GC-MS were recorded on Finnigan Trace
DSQ (Thermo Electron Corporation) at an ionization voltage of
For the characterization and purity epoxidation products of
dienes as follows
7
¼
0 eV equipped with a DB-5 capillary column (internal diameter The precipitate was removed by centrifugation and ltration,
1
0.25 mm, lm thickness ¼ 0.25 mm, length ¼ 30 m). H NMR and the ltrate extracted with EtOAc (30 ml ꢂ 3), the organic
1
3
and C NMR spectra were recorded on a Bruker AM-400 and layer was collected and washed with water and brine, then dried
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 32054–32062 | 32055

View Article Online
RSC Advances
Paper
with anhydrous Na
2
SO
4
. Aer evaporation of the solvent under further conrmed the existence of peroxo n (O–O) band in
reduced pressure, the residue was puried by column chro- compound I.
matography with eluent (petroleum ether/ethyl acetate 10/1) to
To study the solvent effect of the reaction-controlled phase-
give epoxides as colorless liquid. The structures of the epoxides transfer catalytic system, the epoxidation of 1-octene with
1
13
were determined by MS (EI), H NMR and C NMR.
H
(
2
O
2
using catalyst I was carried out at the same temperature
333 K) with different solvents (Table 2). Epoxidation of cyclo-
hexene catalyzed by I with H in ethyl acetate at 333 K
2 2
O
resulted in 99% selectivity and 96% yield. Those solvents of tert-
butanol, acetonitrile, dioxane, toluene, 1,2-dichloroethane and
cyclohexane gave lower yields for the epoxidation of cyclo-
hexene. Although catalyst I was soluble in toluene, 1,2-dichlo-
roethane and cyclohexane during reaction, the catalyst could
not be precipitated out from those solvents when temperature
dropped from 333 K to 298 K. In tert-butanol and dioxane, the
phenomena of solid–liquid–solid phase transfer of catalyst I
existed, but the amount of the catalyst precipitate was very little,
indicating the two solvents were not suitable for recovering
catalyst. Based on the above results, ethyl acetate is the best
candidate solvent. Solvents have effect on forming an efficient
reaction-controlled phase-transfer catalytic system.
The silicotungstate catalyst I is applicable to the epoxidation
of various olens (such as linear terminal olens, internal
2 2
olens, cyclic olens and unactivated alkenes) with 30% H O ,
and high yields and selectivities are achieved (Table 3). The
same phenomena of reaction-controlled phase-transfer catal-
ysis were observed all in these epoxidations, and excellent
catalytic activities were presented. Bulky cyclic olens such as
cyclohexene, 1-methyl-cyclohexene, cyclooctene and norbor-
nene were epoxidized with 99% selectivity and ꢁ90% yield of
epoxides (entry 1–5).
Results and discussion
Epoxidation of various olens
Epoxidation of 1-octene was carried out with various poly-
oxometalates including saturated and lacunary poly-
oxometalates in ethyl acetate at 333 K (Table 1). Compound I
exhibited the highest activity among all the catalysts (entry 1)
and showed a 97% yield of 1,2-epoxyoctane. Saturated silico-
tungstate of [(C H ) (CH ) N] [SiW O ] and saturated tung-
1
8
37 2
3 2
4
12 40
stophosphate of[(C H ) (CH ) N] [PW O ] showed no
1
8
37 2
3 2
3
12 40
activity (entry 2–3). Dioctadecyl dimethyl ammonium salt of the
other lacunary silicotungstates and peroxotungstate showed
poor activity (entry 4–6), and these catalysts were not soluble in
ethyl acetate even at 333 K. As shown in Table 1, no epoxide was
detected for the simple mixture of dioctadecyl dimethyl
ammonium
chloride
(DDAC),
Na
2
WO
4
$2H
2
O
and
Na SiO $9H O, or the mixture of two of them (entry 8–11). The
2
3
2
epoxidation can not proceed without catalyst (entry 12). When
only catalyst I was added without H O , the yield of 1,2-epoxy-
2
2
octane was 2% (entry 13). The producing of 1-octene oxide
should be ascribed to the peroxo species in the catalyst and
Table 1 Epoxidation of 1-octene with H
2
O
2
catalyzed by various
7
Nonactivated terminal C –C16 olens, such as 1-heptene, 1-
octene, 1-undecene, 1-dodecene and 1-hexadecylene, could be
a
catalysts
transformed to the corresponding epoxides specically with
Selectivity
Yield
Entry
Catalyst
I
(mol%)
(mol%) 99% selectivity and ꢁ90% yield of epoxides (entry 12–17). For
epoxidation the gas of propylene, a 94% selectivity and 90%
1
2
3
4
5
6
99
0
0
99
99
99
0
97
0
0
8
8
yield of epoxide (entry 18) were obtained in 2 hours at 333 K, in
[DDA]
[DDA]
[DDA]
[DDA]
[DDA]
4
3
8
8
2
[SiW12
[PW12
[b -SiW11
[g-SiW10
O
40
]
]
O
36
O ]
which the reaction time is far shorter than that of Xi's reaction-
O
40
22
controlled phase-transfer catalyst (5 hours at 338 K). There-
fore, the catalytic performance of this reaction-controlled
2
39
]
[W
3
O
18
]
17
0
0
b
7
[(C18
H
37
)
2
(CH
$9H
$2H
$9H
$2H
$9H
$2H
3
2
) N]Cl
c
8
Na
2
Na
2
Na
2
Na
2
Na
2
Na
2
SiO
WO
3
2
O +
O + DDAC
O +
0
4
2
Table 2 Epoxidation of cyclohexene catalyzed by I under different
solvents
d
9
SiO
WO
3
2
0
0
a
4
2
O
e
10
11
12
13
SiO
WO
3
2
O + DDAC
2
O + DDAC
0
0
0
0
0
0
2
Entry
Solvent
Selectivity (mol%)
Yield (mol%)
f
4
No catalyst
No H
1
2
3
4
5
6
7
8
Ethyl acetate
tert-Butanol
Acetonitrile
Dioxane
99
99
99
99
99
99
99
99
96
84
82
81
78
79
70
45
g
2
O
2
99
a
Reaction conditions: 2.5 mmol 1-octene; 0.5 mmol H
2
O
2
; 10 mmol
catalyst; 2 ml ethyl acetate; reaction temperature: 333 K, reaction
time: 3 h. Selectivity (%) ¼ epoxide (mol)/all products (mol) ꢂ 100.
Toluene
Yield (%) ¼ epoxide (mol)/H
2
O
2
used (mol) ꢂ 100. Conversions and
1,2-Dichloroethane
Cyclohexane
Ethanol
selectivities were determined by gas chromatography using an
2 2
internal standard technique and were based on H O .
b
[
(C18
DDAC), 30 mmol.d 10 mmol Na
0 mmol DDAC. 10 mmol Na
0 mmol Nag
H
37
)
2
(CH
3
)
2
N]Cl ¼ dioctadecyl dimethyl ammonium chloride
c
a
(
2
SiO
SiO
O, 30 mmol DDAC. 20 mmol Na
mmol DDAC. Yield (%) ¼ alkenes' conversion ꢂ epoxide selectivity.
3
$9H
2
O, 30 mmol Na
O, 30 mmol Na
2
WO
4
$2H
2
O,
2 2
Reaction conditions: 2.5 mmol cyclohexene; 0.5 mmol H O ; 10 mmol
3
2
3
$9H
2
2
WO
4
$2H
2
O. catalyst of I; 2 ml solvent; reaction temperature: 333 K, reaction time: 2
e
f
1
2
SiO
3
$9H
2
2
WO
4
$2H
2
O, 60 h. Selectivity (%) ¼ cyclohexene epoxide (mol)/all products (mol) ꢂ 100.
Yield (%) ¼ cyclohexene epoxide (mol)/H
O
2 2
used (mol) ꢂ 100.
Conversions and selectivities were determined by gas chromatography.
32056 | RSC Adv., 2014, 4, 32054–32062
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
a
2 2
Table 3 Epoxidation of various olefins with H O catalyzed by I
Entry
Substrate
Product
Reaction time (h)
Selectivity (mol%)
Yield (mol%)
b
1
2
2
99
99
88
96
2
3
2
99
99
4
5
6
2
1
3
99
99
92
97
99
92
7
2
94
94
8
9
3
2
94
99
94
84
1
0
7
99
65
1
1
1
2
7
2
99
99
62
99
1
1
1
1
3
4
5
6
3
3
3
2
99
99
99
99
98
97
93
92
1
7
8
3
2
99
94
99
90
c
1
a
Reaction conditions: 2.5 mmol alkene; 0.5 mmol H
2
O
2
; 10 mmol catalyst of I; 2 ml ethyl acetate; reaction temperature: 333 K, selectivity (%) ¼
used (mol) ꢂ 100. Conversions and selectivities were determined by gas
epoxide (mol)/all products (mol) ꢂ 100. Yield (%) ¼ epoxide (mol)/H
2 2
O
b
c
chromatography using an internal standard technique and were based on H
2 2 2 2
O . Reaction temperature: 318 K. Propylene 4 atm, 2 ml 30% H O .
phase-transfer catalyst inspires the prospect of using this type a 99% yield). In general, the epoxidation of allylic alcohol and
of catalyst for the industrial epoxidation of propylene instead of allylic chloride are carried out difficultly compared with those of
the classical chlorohydrin process. In the latter status, epoxides cyclic olen, since the C]C double bond in allyl moieties is
are usually consecutively hydrated and the oxirane ring is relatively electron decient by the electron-withdrawing group
cleaved, resulting in a decrease in selectivity. Furthermore, large of hydroxyl (–OH) and –Cl. In our catalytic system, a 84% yield of
amounts of Cl
process, which gives rise to serious problems of equipment obtained over catalyst I (entry 9–10) at 333 K, respectively,
corrosion and environmental pollution. revealing this epoxidation system was effective and fairly
Those olens containing aromatic group, such as styrene, a- universal to all kinds of alkenes.
2
are consumed for the classical chlorohydrin 2,3-epoxy-1-propanol and 65% yield of epichlorohydrin were
methylstyrene and indene, were also converted to correspond-
This new green process of epoxidation is also effective for the
ing epoxides selectively, while a small amount of benzaldehyde epoxidation of dienes (Table 4). Aer two hours reaction, 90%
or corresponding aromatic aldehyde was formed (entry 6–8). conversion for all substrates was obtained, but the selectivity for
Epoxidation rate of 2-heptene was faster than that of 1-heptene, monoepoxide was moderate. Limonene and 4-vinyl-1-cyclohexene,
and the conguration of the C]C bond was retained in the a C]C moiety with higher electron density was epoxidized
corresponding epoxide (cis-2,3-epoxyheptane was obtained with regioselectively to give the corresponding monoepoxide without
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 32054–32062 | 32057

View Article Online
RSC Advances
Paper
a
2 2
Table 4 Epoxidation of dienes with H O catalyzed by I
Entry
1
Dienes
Product
Selectivity (mol%)
99
Yield (mol%)
99
84
2
3
95
30
30
69
69
Fig. 1 Epoxidation of cyclohexene on catalyst I. Reaction conditions: 5
mmol alkene; 1 mmol H
2
O
2
ꢃ
; 20 mmol catalyst of I; 4 ml ethyl acetate;
4
28
28
71
reaction temperature: 60 C, reaction time: 2 h. Conversion (%) ¼
products (mol)/H
2
O
2
used (mol) ꢂ 100. Yield (%) ¼ epoxide (mol)/
2
H O
2
used (mol) ꢂ 100. Selectivity of cyclohexene epoxide was more
than 99% for all cycles.
72
To evaluate the recycling capability of this reaction-
controlled phase-transfer catalytic system, another substrate
of 1-octene that typically inert substrate to be epoxidized was
investigated for epoxidation research (Table S1†). We found the
solubility of silicotungstate catalyst I in 1-octene/ethyl acetate
was better than that in cyclohexene/ethyl acetate, so the loss of
5
49
49
46
46
a
2 2
Reaction conditions: 2.5 mmol alkene; 0.5 mmol H O ; 10 mmol
catalyst of I; 2 ml ethyl acetate; reaction temperature: 333 K, reaction catalyst for the former was more severely than that for the later.
time: 2 h. Selectivity (%) ¼ monoepoxide (mol)/all products (mol) ꢂ The recovery of catalyst in 1-octene/ethyl acetate was not as good
1
00. Yield (%) ¼ epoxide (mol)/H
2
O
2
used (mol) ꢂ 100. Conversions
as that in cyclohexene/ethyl acetate. However, it was exciting
that the yield of epoxide still kept more than 80% even aer
twenty consecutive cycles. No fresh catalyst I was added and
only reaction time was prolonged.
and selectivities were determined by gas chromatography using an
internal standard technique and were based on H
were identied by GC/MS and H NMR.
2
O
2
. The products
1
the successive epoxidation of the other C]C fragment (entry 1–2)
2
1
(
[
similar to these catalysts of [TBA]
4
[g-SiW10
O
34(H
2 2
O) ]
and
8
Kinetics of the epoxidation
p-C H N(CH )15CH ] [PW O32] ). Dicyclopentadiene and geraniol
5 5 2 3 3 4
were displayed moderate selectivity for monoepoxide (selectivity to The kinetics of epoxidation of some typical alkenes with 30%
monoepoxide was ꢁ70%, entry 3–4). When 1-methyl-1,4- hydrogen peroxide catalyzed by I were investigated in the stoi-
cyclohexadiene were epoxidized (entry 5), it gave an equal prod- chiometric condition between the oxidant and the substrates
ucts distribution for the monoepoxide of 3-methyl-7-oxa-bicyclo (Fig. 2). The catalyst I shows best activity for epoxidation of
[4.1.0]hept-3-ene (49% yield) and 1-methyl-7-oxa-bicyclo[4.1.0] cyclooctene, a 99% yield of epoxide was obtained in 1 hour. For
hept-3-ene (46% yield).
2-heptene, it took 4 hours to get the same yield (99%) of 2-
heptene oxide. The initial reaction rate of cyclohexene catalyzed
by I was nearly the same as that of cyclooctene, but the yield of
cyclohexene oxide was 85% in one hour. Aer that, it levelled off
Reuse of catalyst
Another notable feature of this reaction-controlled phase- when the reaction time was prolonged. The initial reaction rate
transfer catalytic system is that the recovery and reuse of the of styrene was far lower than that of cyclohexene, but aer 7
catalyst I was very efficient, easy and convenient. Since the hours reaction, the nal styrene oxide yield (90%) was higher
epoxidation was carried out in solid–liquid–solid phase transfer than that of cyclohexene oxide. The linear terminal olen
mode, the solid catalyst could be easily recovered by ltration or substrate of 1-octene was the most difficult to be epoxidized
centrifugation and reused aer reaction. The recycling experi- among them, even aer 12 hours reaction, only 70% yield of
ments over epoxidation of cyclohexene were shown on Fig. 1. epoxide was obtained. All the epoxidations catalyzed by I
The yield of cyclohexene epoxide did not drop even aer twenty showed no induction period, which is different to the poly-
consecutive cycles, a 90% yield of the epoxide with 99% selec- oxometalate of [g-SiW O (H O) ] that showed an induction
1
0
34
2
2
tivity was still kept for the epoxidation of cyclohexene. No fresh period of approximately 60 min during the epoxidation of
21
catalyst I was added and no reaction time was prolonged.
cyclooctene.
32058 | RSC Adv., 2014, 4, 32054–32062
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
recycled catalysts indicates that the structure of dioctadecyl
dimethyl ammonium salt of silicotungstate catalyst II has not
changed and it was stable during the catalytic cycles from the
2
9
second catalytic cycle. Furthermore, the Si MAS NMR spectra
of fresh and used catalysts are nearly the same (Fig. 4c and d),
indicating the catalyst keeps stable from second cycle. So, the
silicotungstate I should be called pre-catalyst accurately. The
used catalyst aer rst run is the true catalyst and is named as
silicotungstate II. The high activity was maintained for many
cycles should be attributed to the stable structure of the recycled
catalysts.
2
9
The Si MAS NMR spectrum of silicotungstate I showed one
signal at ꢀ79.2 ppm (Fig. 4b), which is different to that of
known silicotungstates, such as K
8
[g-SiW10
O
36] (ꢀ85.0 ppm)
Fig. 2 Reaction profiles for the epoxidation of different olefins with
one equivalent of 30% hydrogen peroxide catalyzed by I. Reaction
29
(Fig. 4a) and H
4
[SiW12
O
40] (ꢀ84.4 ppm), suggesting that I is a
new pure silicotungstate compound.
2 2
conditions: 0.5 mmol alkene; 0.5 mmol H O ; 20 mmol catalyst of
[
(C18 (CH N] [SiO H(WO ]; ml ethyl acetate; reaction
H
37
)
2
3
)
2
3
4
5
)
3
2
The NMR signal of this silicotungstate complex moves
temperature: 333 K. Conversions and selectivity were determined by toward downeld and is similar to the wellknown Venturello
gas chromatography using an internal standard technique.
3ꢀ
3ꢀ
peroxocomplex of {PO [WO(O ) ] } (or [PW O ] ), derived
4
2 2 4
4
24
from the decomposition of Q [PW O ], which also shis from
3
12 40
14
upeld (ꢁꢀ14 to 15 ppm) to downeld (ꢁ3–4 ppm).
2
9
While the Si MAS NMR spectrum of used catalyst (II) gives
several broad peaks whose chemical shi moves toward upeld,
ranging from ꢀ115 ppm to ꢀ130 ppm (Fig. 4c and d). The
original single peak at ꢀ79.2 ppm for I has disappeared. The
new broad peaks for the used catalyst II suggest that the catalyst
is a mixture that contains several different silicotungstate
Characterization analysis
The FT-IR spectrum of the fresh catalyst I (Fig. 3a) gave bands at
ꢀ
1
1000, 967, 942, 846, 722, 634, 574, 511, 473 and 448 cm in the
ꢀ
1

ngerprint region. The IR of 574, 846, 942 and 967 cm can be
assigned to n (W(O )), n (O–O), n (Si–OH) and n (W]O) [20, 26],
2
ꢀ
1
2
9
respectively. The presence of 846 cm proves that there exists
peroxo n (O–O) band in the compound of silicotungstate I.
Compared with the IR spectra of fresh catalyst and used one, the
bands of them are different and the used catalyst has converted
into another species (named as silicotungstate II). However,
since the catalyst began to recycle, the IR spectra of the rst,
second, third and h recycled ones show nearly the same
bands (Fig. 3b–e). The similarity of IR spectra among the
species. It can be seen that the Si MAS NMR spectra of used
catalyst for cycle 1 and cycle 2 are nearly same (Fig. 4c and d),
revealing that the structure of these used catalysts are same and
the catalyst keeps stable from the catalytic cycles.
1
83
The
3
W NMR spectrum of I in CDCl shows one signal at
ꢀ
316 ppm (Fig. 5), which is different to that of other reported
30
catalytically peroxotungstate, such as [(n-C H ) N] [{WO(O ) }-
4
9 4
2
2 2
2(
m-O)] (ꢀ587.5 ppm) and silicotungstate of (Bu N) [g-
4
4
21
183
SiW O (H O) ]
(
W NMR showed ve signals at d ¼ ꢀ95.7,
1
0
34
2
2
2
9
Fig. 3 FT-IR spectra of (a) the fresh catalyst of catalyst I; (b) the used Fig. 4
Si MAS NMR spectra of catalysts: (a) K
(CH N] [SiO H(WO ]; (c) the catalyst of cycle 1; (d) the
catalyst of cycle 2.
8 2
[g-SiW10O36]$12H O;
catalyst for cycle 1; (c) the used catalyst for cycle 2; (d) the used (b) [(C18
H
37
)
2
3
)
2
3
4
5 3
)
catalyst for cycle 3; (e) the used catalyst for cycle 5.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 32054–32062 | 32059

View Article Online
RSC Advances
Paper
183
Fig. 5
3
W NMR spectrum of catalyst I (16.6 MHz, CDCl , 300 K).
ꢀ
98.9, ꢀ118.2, ꢀ119.6, and ꢀ195.7 ppm). The single signal of
1
83
W NMR of I indicates that the chemical environment of the
three tungsten atoms in compound I is identical.
The elemental analysis of I showed that the dioctadecyl
dimethyl ammonium : Si : W mol ratio is 3 : 1 : 3. The UV-vis
spectrum of I in CHCl
3
shows an absorption band at 243 nm
(
Fig. S3†), which is the characteristic band of the peroxotung-
31
b
state. The characteristic IR band of the n (W–O –W) (corner-
sharing) and n (W–O –W) (edge-sharing) in polymerization of
c
W species of Keggin polyoxometalate should be in the range of
ꢀ
1
32
870–880 and 780–790 cm
,
respectively. However, those Scheme 2 Reaction mechanism for the epoxidation of olefins with
bands are not observed in the FT-IR spectrum of I, suggesting
the absence of polymerization of W species in catalyst I. The
2 2
H O catalyzed by I.
33
ꢀ1
characteristic band of n (Si–OH) at 942 cm is observed in the
IR spectrum of I. Based these results, we proposed a structure of epoxidation catalyzed by III can proceed successively until the
I may be [(C18 (CH N] [(OH)SiO {WO(O
] (Scheme 1), use up of oxidant. These soluble species III lost active oxygen
which is similar to the well known Venturello complex and then polymerized into larger silicon-containing species II
H
37
)
2
3
)
2
3
3
2 2 3
) }
3
ꢀ
of[{W (]O) (O ) } (m-PO )] . Unfortunately, even tried our which was insoluble in ethyl acetate and water at room
2
2 4
4
best, we did not obtain crystallographic quality single crystals of temperature. Thus, the used catalyst of silicon-containing
catalyst I in various solvents with vapor diffusion of poor species II precipitated out from the reaction solution aer the
solvents (such as diethyl ether and methanol) so far. Therefore, hydrogen peroxide was consumed up with the decreasing of
the precise structural formula of I can not be given so far.
reaction temperature. The used catalyst II could be recovered by
ltration or centrifugation of the above solid–liquid mixture.
The changes of structure for pre-catalyst I and catalyst II could

Mechanism of the epoxidation of olen by I
2
9
be claried by the IR and Si MAS NMR characterization.
For the recycled reactions, rstly, the insoluble catalyst of
silicon-containing species II was transferred to the active
silicon-containing species III under the action of H O . Then, a
On the basis of all the results, we propose a reaction mechanism
for the epoxidation of olens with hydrogen peroxide catalyzed
37 2 3 2 3
by I (Scheme 2). First, when pre-catalyst of [(C18H ) (CH ) N] -
2
2
[
SiO H(WO ) ], alkenes and H O were mixed in solvent, the
4
5 3
2 2
similar epoxidation process as above proceeded. Aer reaction,
the active silicon-containing species III lost its active oxygen and
became the silicon-containing species II again, and could be
precipitated from the reaction solvent. The used catalyst II was
recovered by ltration or centrifugation from the solid–liquid
mixture. Therefore this reaction-controlled phase-transfer
epoxidation catalyst was recycled many times based on the
above reaction mechanism.
pre-catalyst reacts with hydrogen peroxide to form another
silicon-containing species III that contains the peroxo group.
With the increasing of reaction temperature and the aid of long
chain of the quaternary ammonium cation, the active silicon-
containing species III gradually dissolved in the solvent while
the epoxidation reaction took place. The active oxygen in III was
transferred to C]C bond of the olens to afford the epoxide.
Once the active oxygen of compound III was losing, it will
regenerate again immediately by the oxygen reservoir of the
oxidant of hydrogen peroxide in the reaction system. So, the
Although the silicon-containing species II is the true catalyst,
it could not be synthesized directly so far. This case was just as
Xi's famous reaction-controlled phase-transfer catalyst [p-
C
5
H
5
2 3 3 4
N(CH )15CH ] [PW O16], which also acted as precursor for
the real catalytic cycles. The recyclable solid catalyst was a
3
ꢀ
7ꢀ
11 39
mixture of [PW O ] , [PW O ] and some unidentied
1
2
40
23
heteropolyoxotungstates.
Oxidation of suldes
The present system also catalyzes the oxidation of various
suldes to sulfoxides and sulfones in reaction-controlled phase-
transfer mode. The results are summarized in Table 5. Thio-
anisole, benzothiophene and dibenzothiophene (entry 1, 6 and
Scheme
ꢀ
2 2 3
SiO {WO(O ) } ] .
1 Proposed structure of the anion of I: [(OH)
3
3
32060 | RSC Adv., 2014, 4, 32054–32062
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
a
Table 5 Oxidation of various sulfides with hydrogen peroxide catalyzed by I
Entry
1
Substrate
Product 1
Product 2
Conversion (%)
100
Yield 1 (%)
0
Yield 2 (%)
100
2
3
59
64
35
58
24
6
4
5
67
45
22
91
81
10
b
6
100
—
100
c
7
100
—
100
a
2 2
Reaction conditions: time 2 h; 2 ml ethyl acetate; 10 mmol (25 mg) catalyst of I; 1 mmol substrate; 2 mmol H O ; reaction temperature: 333 K; yield
(
%) ¼ products (mol)/substrate used (mol) ꢂ 100. Conversions and selectivity were determined by gas chromatography using an internal standard
b
c
technique were based on the substrate. Reaction temperature: 343 K, 4 ml solvent ethyl acetate. Reaction temperature: 353 K. 4 ml ethyl acetate.
7) were selectively mono-oxygenated to the corresponding
The recovery and reuse of the catalyst I was also very efficient,
sulfones with 100% conversion and 100% yields. For thiophene easy and convenient in the oxidation of suldes. The same
and its methyl-substituted derivatives, the conversions of phenomena of reaction-controlled phase-transfer catalysis were
substrates were not completed in 2 hours and both sulfoxides observed all in these oxidations and excellent catalytic activity was
and sulfones were produced. Whereas 2,5-dimethylthiophene presented. The recycling experiments over oxidation of thio-
was selectively mono-oxygenated to the corresponding sulfox- anisole were shown on Table S2.† The yield of methyl(phenyl)
ides (ca. 81%) along with formation of the corresponding sulfane did not drop aer ve consecutive cycles, a 99% yield of
sulfones (ca. 10%).
the sulfone was still kept for the oxidation of thioanisole.
a
2 2
Table 6 Oxidation of various alcohols catalyzed by I with H O
Entry
1
Substrate
Product
Conversion (mol%)
76
Selectivity (mol%)
100
Yield (mol%)
76
2
3
45
100
45
48
87
100
100
48
87
4
a
Reaction conditions: 10 mmol catalyst; 2 ml ethyl acetate; 1 mmol substrate; 30% aq. H
2 2
O 5 mmol; reaction temperature: 353 K, reaction time:
1
0 h; the conversion was based on alcohols, the selectivity was based on ketone, aldehyde or acid.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 32054–32062 | 32061

View Article Online
RSC Advances
Paper
Oxidation of alcohols
8 Y. Ding, W. Zhao, H. Hua and B. Ma, Green Chem., 2008, 10,
9
10–913.
R. Neumann and M. Gara, J. Am. Chem. Soc., 1994, 116, 5509–
510.
Different kind of alcohols were converted into the correspond-
ing ketones under this reaction-controlled phase transfer cata-
lytic system. The results in Table 6 show that I effectively
catalyzes the oxidation of alcohols to corresponding carbonyl
compounds with hydrogen peroxide. Normal to high conver-
sions and high selectivities for secondary alcohols were
observed under our reaction conditions. Especially, the 2-octa-
nol showed a 87% conversion and a 87% yield for octan-2-one.
9
5
1
1
0 Y. Ding, Q. Gao, G. Li, H. Zhang, J. Wang, L. Yan and J. Suo, J.
Mol. Catal. A: Chem., 2004, 218, 161–170.
1 L. Salles, C. Aubry, R. Thouvenot, F. Robert, C. Doremieux-
Morin, G. Chottard, H. Ledon, Y. Jeannin and
J. M. Bregeault, Inorg. Chem., 1994, 33, 871–878.
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
2 N. M. Gresley, W. P. Griffith, A. C. Laemmel, H. I. S. Nogueira
and B. C. Parkin, J. Mol. Catal. A: Chem., 1997, 117, 185–198.
3 K. Kamata, K. Yonehara, Y. Nakagawa, K. Uehara and
N. Mizuno, Nat. Chem., 2010, 2, 478–483.
4 D. C. Duncan, R. C. Chambers, E. Hecht and C. L. Hill, J. Am.
Chem. Soc., 1995, 117, 681–691.
5 C. Venturello, R. D'Aloisio, J. C. Bart and M. Ricci, J. Mol.
Catal. A: Chem., 1985, 32, 107–110.
6 C. Venturello and R. D'Aloisio, J. Org. Chem., 1988, 53, 1553–
Conclusions
In summary, a new reaction-controlled phase-transfer catalyst
based on silicotungstate of [(C18H ) (CH ) N] [SiO H(WO ) ] (I)
37 2 3 2 3 4 5 3
was developed. Complex I exhibited high catalytic activity for
epoxidation of various alkenes (such as linear terminal olens,
internal olens, cyclic olens, styrene, and unactivated alkenes) to
epoxides, the oxidation of suldes and the oxidation of alcohols.
This new reaction-controlled phase-transfer catalytic system
not only made the recovery of catalyst easy, but also avoided the
using of toxic solvents. All the oxidation reactions were per-
formed in a recyclable and environmentally benign catalytic
1557.
7 H. Yamamoto, M. Tsuda, S. Sakaguchi and Y. Ishii, J. Org.
Chem., 1997, 62, 7174–7177.
8 K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi,
S. Hikichi and N. Mizuno, Science, 2003, 300, 964–966.
9 Y. Nakagawa, K. Kamata, M. Kotani, K. Yamaguchi and
N. Mizuno, Angew. Chem., Int. Ed. Engl., 2005, 44, 5136–5141.
0 K. Kamata, Y. Nakagawa, K. Yamaguchi and N. Mizuno, J.
Catal., 2004, 224, 224–228.
1 K. Kamata, M. Kotani, K. Yamaguchi, S. Hikichi and
N. Mizuno, Chem.–Eur. J., 2007, 13, 639–648.
2 X. Zuwei, Z. Ning, S. Yu and L. Kunlan, Science, 2001, 292,
system: [(C18
substrates. The catalyst was easily recovered aer reaction by
ltration and centrifugation, and the high activity was main-
tained for the recycling reactions. It was found that silicotung-
state of [(C18 (CH N] [SiO H(WO ] (I) was a pre-catalyst
37 2 3 2 3 4 5 3 2 2
H ) (CH ) N] [SiO H(WO ) ]/H O /ethyl acetate/

H
37
)
2
3
)
2
3
4
5 3
)
and the true catalyst was a mixture that contains several
different silicotungstate species.
1139–1141.
Acknowledgements
3 J. Gao, Y. Chen, B. Han, Z. Feng, C. Li, N. Zhou, S. Gao and
Z. Xi, J. Mol. Catal. A: Chem., 2004, 210, 197–204.
4 N. Zhou, Z. Xi, G. Cao and S. Gao, Appl. Catal., A, 2003, 250,
This work was nancially supported by the National Natural
Science Foundation of China (Grant nos 21172098, 21173105,
239–245.
20803032) and the Fundamental Research Funds for the Central
5 A. Mouret, L. Leclercq, A. Muhlbauer and V. Nardello-Rataj,
Green Chem., 2014, 16, 269–278.
6 Y. Ding, B. Ma, D. Tong, H. Hua and W. Zhao, Aust. J. Chem.,
Universities (lzujbky-2013-50 and lzujbky-2014-67).
Notes and references
2009, 62, 739–746.
7 Y. Ishii, K. Yamawaki, T. Ura, H. Yamada, T. Yoshida and
M. Ogawa, J. Org. Chem., 1988, 53, 3587–3593.
8 Y. Ding, Q. Gao, G. Li, J. Wang, L. Yan and J. Suo, Acta Chim.
Sin., 2005, 63, 1167.
1
A. Rezaeifard, R. Haddad, M. Jafarpour and M. Hakimi, J. Am.
Chem. Soc., 2013, 135, 10036–10039.
2
W. Lueangchaichaweng, N. R. Brooks, S. Fiorilli,
E. Gobechiya, K. Lin, L. Li, S. Parres-Esclapez, E. Javon,
S. Bals and G. Van Tendeloo, Angew. Chem., Int. Ed., 2014,
9 A. Oulmekki and F. Lefebvre, React. Kinet. Catal. Lett., 1992,
48, 601–606.
53, 1585–1589.
0 K. Kamata, S. Kuzuya, K. Uehara, S. Yamaguchi and
N. Mizuno, Inorg. Chem., 2007, 46, 3768–3774.
1 C. Aubry, G. Chottard, N. Platzer, J. M. Bregeault,
R. Thouvenot, F. Chauveau, C. Huet and H. Ledon, Inorg.
Chem., 1991, 30, 4409–4415.
3
4
M. De Torres, I. W. Arends, J. A. Mayoral, E. Pires and
G. Jim ´e nez-Os ´e s, Appl. Catal., A, 2012, 425, 91–96.
S. Doherty, J. G. Knight, J. R. Ellison, D. Weekes,
R. W. Harrington, C. Hardacre and H. Manyar, Green
Chem., 2012, 14, 925–929.
3
2 T. Okuhara, N. Mizuno and M. Misono, Adv. Catal., 1996, 41,
113–252.
3 D. Xu, L. Sun, H. Li, L. Zhang, G. Guo, X. Zhao and L. Gui,
5
6
B. Notari, Adv. Catal., 1996, 41, 253–334.
R. Llamas, C. Jim ´e nez-Sanchidri ´a n and J. R. Ruiz, Appl.
Catal., B, 2007, 72, 18–25.
3
New J. Chem., 2003, 27, 300–306.
7
M. Timofeeva, Z. Hasan, A. Y. Orlov, V. Panchenko,
Y. A. Chesalov, I. Soshnikov and S. H. Jhung, Appl. Catal.,
B, 2011, 107, 197–204.
32062 | RSC Adv., 2014, 4, 32054–32062
This journal is © The Royal Society of Chemistry 2014