A role of ionic liquid as an activator for efficient olefin epoxidation catalyzed by polyoxometalate

Article abstract of DOI:10.1039/b710444h

Ionic liquid (IL), such as 1-butyl-3-methylimidazolium hexafluorophosphate ([bmim][PF₆]), as an activator for the epoxidation of a wide variety of olefins with H₂O₂ catalyzed by Keggin polyoxometalate [bmim]₃PW<

Full text of DOI:10.1039/b710444h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
A role of ionic liquid as an activator for efficient olefin epoxidation
catalyzed by polyoxometalate
a
a
a
a
a
Lili Liu, Chuncheng Chen, Xuefeng Hu, Tariq Mohamood, Wanhong Ma,
b
a
Jun Lin and Jincai Zhao*
Received (in Montpellier, France) 9th July 2007, Accepted 14th September 2007
First published as an Advance Article on the web 24th September 2007
DOI: 10.1039/b710444h
6
Ionic liquid (IL), such as 1-butyl-3-methylimidazolium hexafluorophosphate ([bmim][PF ]),
as an activator for the epoxidation of a wide variety of olefins with H
2 2
O catalyzed by Keggin
polyoxometalate [bmim] 40 has been demonstrated. Use of IL results both in significant
3
PW12O
enhancement of TOF and selectivity for olefin epoxidation compared to that of traditional
organic solvents, e.g. 289 times TOF and 1.3 times selectivity as found in CH Cl for the
2
2
3
epoxidation of cis-cyclooctene. The kinetic P NMR spectra uncover that the active
1
3
À
peroxotungstate [PO
4
{W(O)(O)
}
2 4
]
(Venturello complex), contributing to the epoxidation,
]. Furthermore, the counterion such
can be readily generated in the presence of [bmim][PF
6À
6
À
as PF and [(CF
3
SO
2
)
2
N] is evidenced to play a crucial role in the epoxidation of olefin.
4À
6À
BF as counterion, instead of PF , is found to reduce TOF and selectivity for
cis-cyclooctene epoxidation dramatically from 289 to less than 1 and from 99% to 12%,
respectively.
Introduction
it has been extensively investigated in other media. Moreover,
some studies show that, besides as ‘‘green’’ reaction media,
ionic liquids seem to play some special roles in catalytic
reactions. For example, Song and Roh showed that the
addition of ionic liquid could clearly accelerate the epoxida-
tion of 2,2-dimethylchromene in Jacobsen’s chiral (salen)Mn
In recent years, tungsten-based polyoxometalates have re-
ceived an increasing interest in selective oxidation of organic
compounds with hydrogen peroxide, i.e. Ishii–Venturello
chemistry, due to their high efficiency and the use of the
1
–10
environmentally friendly oxidant.
However, a series of
15
catalytic system. Sun et al. reported that an ionic liquid was
drawbacks, for example, the use of toxic and carcinogenic
chlorinated solvents and the difficulty in reuse of catalyst and
separation of products in traditional biphasic or homogeneous
systems, diminish their advantages.
beneficial for the oxidation of alcohols in a copper(II) chloride
21
catalytic system. Hardacre and co-workers reported the
superior rates and selectivities for the sulfoxidation in ionic
liquids compared with molecular solvents with catalysts Ti-
22
Much recent research in catalysis is associated with manip-
ulation and optimization of reaction media to achieve high
SBA-15 and UL-TS-1. Mizuno and co-workers found that
peroxotungstate immobilized by an ionic liquid on SiO
2
1
1,12
efficiency and to decrease the environmental load.
Ionic
particles exhibited superior activity with respect to that of
23
triethylammonium cation for the epoxidation of olefins.
liquids, which possess many desirable properties such as
negligible vapor pressure, high thermal stability, good sol-
vency of organic and inorganic materials and the possibility to
influence the chemical reaction process, are usually used as
However, the mechanism underlying the enhancement in
reactivity of ionic liquids is not clear.
We herein exploit the hydrophobic 1-butyl-3-methylimida-
‘
‘green’’ and recyclable media for chemical processes including
zolium ionic liquid (e.g. [bmim][PF ]) as a solvent for epoxi-
6
organometallic catalysis, electrochemistry and organic synth-
3–19
dation of olefins by the imidazolium-based Keggin polyoxo-
metalate ([bmim] PW O ) with aqueous hydrogen peroxide.
1
esis.
However, only a few examples of polyoxometalates
3
12 40
used as catalyst in ionic liquids have been reported. For
example, Tandon and co-workers reported the oxidation of
alcohols with hydrogen peroxide catalyzed by a imidazolium
ion based peroxotungstophosphate in the ionic liquid
It was found that [bmim] PW O could epoxidize olefins
3 12 40
with aqueous hydrogen peroxide at a much more rapid rate in
ionic liquids than that in classical organic solvents. The
3
epoxidation reaction in CH CN was also greatly promoted
2
0
[
bmim][BF4]. However, no polyoxometalate-catalyzed epox-
upon the addition of the ionic liquid. The unique role of the
ionic liquid as a cocatalyst in olefin epoxidation catalyzed by
polyoxometalate was demonstrated based on kinetics, solvent
effect, ion effect and spectroscopic experiments. It is suggested
that the ionic liquid could not only be used as solvent for the
reuse of catalyst, but also supply the special microenvironment
to facilitate the formation of the active peroxotungstate from
[bmim] PW O and H O .
idation of olefins in ionic liquids was reported so far, although
a
Key Laboratory of Photochemistry, Beijing, National Laboratory,
for Molecular Sciences, Institute of Chemistry, The Chinese
Academy of Sciences, Beijing, China. E-mail: jczhao@iccas.ac.cn;
Fax: 86 10 82616495; Tel: 86 8261 0080
Department of Chemistry, Renmin University of China, Beijing,
China
b
3
12 40
2
2
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
New J. Chem., 2008, 32, 283–289 | 283

View Article Online
Experimental
Transfer Infrared Spectrometer (Tekmar Dohrmann). The
ionic liquids were characterised using ESI-MS (APEX a model
FT-TCR mass spectrograph). A Bruker 300 MHz instrument
and Bruker 400 MHz instrument were used for NMR spectra.
Materials and catalysts
cis-Cyclooctene, cyclohexene, 1-octene, styrene, a-methylstyr-
ene, trans-2-octene, 2-methyl-1-heptene, 2-methyl-2-heptene
and trans-b-methyl were purchased from Aldrich and Across
Co. and were used as received. Hydrogen peroxide (30 wt%),
3
1
Chemical shifts of P NMR were referenced with respect to
external 85% H PO .
3
4
3
Keggin polyoxometalate (H PW12O40) and tetrabutylammo-
Results and discussion
nium bromide were obtained from Beijing Chemicals Co. All
other chemicals were of laboratory reagent grade quality and
were used without further purification. 1-Butyl-3-methylimi-
dazolium chloride salt ([bmim]Cl) was prepared as described
The activity of catalyst for epoxidation of olefins (Scheme 1)
under different conditions is shown in Table 1. For the
epoxidation of cis-cyclooctene, the catalytic reaction in the
2
4
previously.
1-Butyl-3-methylimidazolium hexafluoropho-
ionic liquid [bmim][PF ] exhibited both a high yield of 87%
6
sphate ([bmim][PF ]) was prepared with [bmim]Cl and
6
and a high selectivity of 99% (entry 1), and the efficiency of
27
KPF : [bmim]Cl (0.1 mol) and KPF (0.1 mol) were added
6
6
H O utilization was determined as ca. 87%. In the blank
2
2
into distilled water. The obtained ionic liquid organic phase
was washed thoroughly with distilled water, and then dried for
experiments, no reaction occurred in the absence of oxidant
or [bmim] 40. Other olefins, including other cyclic
olefins, long-chain olefins and olefins with electron-withdraw-
ing groups, could be efficiently epoxidized in the [bmim]
/[bmim][PF ] system (entries 2–7). Notably, 2-
H
2
O
2
3
PW12O
2
4
1
h using anhydrous MgSO
4
;
d
H
(400 MHz; acetone-d
6
;
Me
4
Si) 0.92 (3H, t), 1.32 (2H, m), 1.80 (2H, m), 3.82 (3H, s),
3
-
4
.12 (2H, t), 7.35 (1H, s), 7.38 (1H, s) and 8.45 (1H, s).
PW12
O40/H
2
O
2
6
Similarly, 1-butyl-3-methylimidazolium bis(trifluoromethane-
2
octene could be readily epoxidized into its epoxide, while 1-
octene was oxidized at rather low rate. As is known, epoxida-
tion is an electrophilic reaction, and the methyl moiety is a
good electron-donating group. The high activity of 2-octene
compared to that of 1-octene could be ascribed to the effect of
electron-donating methyl substitution. Similarly, styrene with
an electron-withdrawing aromatic group was epoxidized with
rather low yield and selectivity. However, trans-b-methylstyr-
ene with an electron-donating methyl group exhibited an
increased reactivity. Another important fact for the present
catalytic system is that the configuration around the double
bond of olefins was completely retained in the corresponding
epoxides in the epoxidation of cis- and trans-olefins. The
stability and the recyclable of the [bmim] PW O /
5
3
sulfonyl)imide ([bmim][NTf ]) and 1-butyl-3-methylimidazo-
2
4
lium tetrafluoroborate ([bmim][BF4]) were prepared by the
method of ion exchange. [bmim] PW O was precipitated
3
12 40
from aqueous solutions of H PW O by the addition of
3
12 40
2
excess [bmim]Cl. The white solids were collected, washed
6
thoroughly with H
2
O, and dried under vacuum overnight;
À1
n
1
5
max(film)/cm
3441, 3149, 3116, 2961, 2932, 2870, 1636,
563, 1459, 1384, 1164, 1079, 979, 895, 806, 747, 650, 621,
2 3 3 40
21. [Bu N] 40 and [C N(CH )15CH ] PW12O
4
3
PW12O
H
5 5
were prepared with tetrabutylammonium bromide and cetyl-
pyridinium bromide, respectively, by a similar method.
Oxidation of olefins
3
12 40
In a typical reaction, the catalyst [bmim]
.4 mmol) was dissolved in [bmim][PF ] (1 cm ). Substrate
0.8 mmol) was added and followed by adding of hydrogen
3
PW12
O
40 (8 mg,
[bmim][PF ] catalytic system was examined by repetitive use
6
3
2
6
of the catalyst and the ionic liquid. It was found that five
recycles for epoxidation of cis-cyclooctene led to no significant
decrease in reactivity and selectivity.
(
peroxide (30% in water) (1.2 mmol). The reaction mixture was
stirred at 60 1C for 1 h. After completion of the reaction, the
Other solvents were also examined for the epoxidation
reaction (Table 1, entries 8–12). Another ionic liquid
3
reaction mixture was extracted with diethyl ether (3 Â 3 cm ).
The extract layer was collected and then a definite amount of
toluene was added as an internal standard for GC analysis.
The ionic liquid phase including the catalyst was washed with
diethyl ether and dried prior to being recycled. The same
reaction conditions were used in other solvents. For classical
3 2 2
[bmim][(CF SO ) N] exhibited similar yield (82%) and selec-
tivity (94%). Interestingly, the ionic liquid [bmim][BF
À
4
] with
[BF
4
]
as counterion showed very low catalytic activity. The
influence of the possible impurities in the ionic liquid was
checked. Halides, such as chloride or fluoride ions, and acid,
are considered as the most possible impurities, which are
derived from the preparation of the ionic liquid and hydrolysis
organic solvents, the reaction mixture was diluted to about
3
cm with diethyl ether after reaction for 1 h and then toluene
9
À À
of the [BF ] or [PF ] , respectively. No notable decrease in
4 6
was added as an internal standard for GC analysis.
the reaction activity was observed when hydrochloric acid
and/or hydrofluoric acid (r1 Â 10À3 M) were added to the
Instrumentation and methods
The epoxidation yield was determined by GC using a gas
chromatograph Hitachi G-3900 equipped with a flame ioniza-
tion detector and a 30 m  0.25 mm DB-5 capillary column.
The oxidation products were identified by means of GC-MS
and GC using authentic samples. For GC-MS analyses, a
Finnigan Trace GC ultra gas chromatograph equipped with a
Finnigan TraceDSQ mass spectrometer was used. The IR
spectra were recorded in KBr on a DigiLab FTS 3500 Fourier
Scheme 1 Epoxidation of olefins in the ionic liquid [bmim][PF6].
2
84 | New J. Chem., 2008, 32, 283–289
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
a
Table 1 Epoxidation of olefins with H
2
O
2
catalyzed by [bmim]
3
PW12
O
40
b
Yield (%)
c
Selectivity (%)
d
Substrate
Solvent
Product
TON
1
2
3
4
5
6
7
8
9
1
1
1
a
cis-Cyclooctene
Cyclohexene
1-Octene
trans-2-Octene
2-Methyl-2-heptene
Styrene
trans-b-Methylstyrene
cis-Cyclooctene
cis-Cyclooctene
cis-Cyclooctene
cis-Cyclooctene
cis-Cyclooctene
[bmim][PF
[bmim][PF
[bmim][PF
[bmim][PF
[bmim][PF
[bmim][PF
[bmim][PF
6
6
6
6
6
6
6
]
]
]
]
]
]
]
Cyclooctene oxide
Cyclohexene oxide
1,2-Epoxyoctane
trans-2,3-Epoxyoctane
2-Methyl-2,3-epoxyheptane
Styrene oxide
trans-b-Methylstyrene oxide
Cyclooctene oxide
Cyclooctene oxide
Cyclooctene oxide
Cyclooctene oxide
Cyclooctene oxide
3
87
36
12
37
67
9
22
99
80
86
94
89
32
67
94
12
77
40
75
289
121
40
125
223
31
74
[bmim][(CF
[bmim][BF
3
SO
2
)
2
N]
82
Trace
1
Trace
Trace
274
o1
3
1
1
4
]
0
1
2
CH
CH
CH
3
3
2
CN
OH
Cl
2
b
2 2
Reaction conditions: 0.8 mmol substrate; 1.2 mmol H O ; 2.4 mmol catalyst; 1 cm solvent; reaction temperature 60 1C; reaction time 1 h. Yield
d
c
for epoxides based on substrate. Selectivity for epoxides determined by GC analysis. Turnover number (TON) = mol products per mol
catalyst.
most effective catalytic system ([bmim]
PF ]–H ). Besides, the presence of water is another poten-
tial problem for [bmim][BF ] ionic liquid system due to its
3
PW12
O
40–[bmim]-
negligible promotive effect, which indicates again that the
anion, besides the cation, in the ionic liquid presents important
effects on the activity of catalyst.
[
6
2 2
O
4
3
1
good aqueous solubility. In this study, the dried ionic liquid
and the optimum amount of aqueous H O for the catalytic
P NMR spectra have proved to been a strong tool for the
3–5,28,29
12 40
3
determination of P species derived from PW O .
1
2
2
activity were used to decrease the influence on the activity
efficiently. It is obvious that the effect of the possible impu-
rities can be excluded under the same experimental conditions.
Therefore, the anion of the ionic liquid may play a key role in
However, most of the reported experiments were carried out
under static conditions, i.e. the NMR spectra were recorded at
3À
different PW12O40 : H O ratios after a given reaction time.
2 2
3
1
In the present study, the time-dependent P NMR spectra of
the polyoxometalate species during the reaction were moni-
tored to reveal their evolvement process. As shown in Fig. 3,
the [bmim]
tested classical organic solvents, including CH
and CH OH, only little cyclooctene oxide was produced under
otherwise identical conditions. In CH Cl , in which the
bmim] 40 is insoluble, a trace of cyclooctene oxide
3
PW12
O
40-based selective oxidation process. In all
2
Cl , CH CN
2
3
3
after [bmim]
3
PW12
O
40 was treated with 50 equivalents of
] (1 : 1) at
2
2
H
2
O
2
in the mixed solvent acetonitrile–[bmim][PF
6
3
À
[
3
PW12
O
50 1C for 15 min, the peak for the original PW12
O
40 (À14.4
was produced. Even in the miscible solvents CH CN and
3
ppm) became very low. Concomitantly, new species with
resonances at 4.1, À12.0 and À12.1 ppm appeared
(Fig. 3(A), curve b). The peak at 4.1 ppm has been assigned
CH OH, the reactivity for the epoxidation was also rather
3
low, which suggests that the solubility was not the only cause
for the high yield of epoxides in ionic liquids.
3À
to be the catalytically active species [PO {W(O)(O) } ]
4 2 4
In order to illuminate the effect of ionic liquid clearly, a
6
mixed solvent of [bmim][PF ] and acetonitrile was used in the
epoxidation of cis-cyclooctene (Fig. 1). Epoxidation reaction
was rather slow (less than 1%) when only acetonitrile was used
6
as solvent. However, when [bmim][PF ] was added to acetoni-
trile to form a homogeneous solution, the catalytic activity was
markedly enhanced. After addition of 10% of ionic liquid,
about 50% of initial cis-cyclooctene was epoxidized to
cyclooctene oxide in 1 h. When the content of ionic liquid in
the mixed solvent was more than 30%, the disappearance of
the cyclooctene and the yield of epoxide tended to remain
constant. It is notable that the total mass was well-balanced
for all the ratios, indicating the good selectivity of the mixed
solvent system.
Tetraalkylammonium salts have been widely used as phase
transfer agents in polyoxometalate-based catalysts. In the
+
À
present study, different types of organic ion salts (C A ) were
also used to determine the effect of the ions (Fig. 2). Results
indicated that neither the short-chain tetrabutylammonium
bromide ([Bu N]Br) nor the long-chain cetyltrimethylammo-
4
6
Fig. 1 Effect of [bmim][PF ] in mixed solvents (acetonitrile and ionic
nium bromide ([CH (CH ) (CH ) N]Br) exhibited much ac-
3
2 15
3 3
liquid) on the epoxidation of cis-cyclooctene at 60 1C for 1 h: (’)
amount of cis-cyclooctene remaining (%); (J) yield of cyclooctene
oxide (%); (,) the total mass balance (%). [Catalyst] = 2.4 mM,
[substrate] = 0.8 M and [H O ] = 1.2 M.
celerative effect on the epoxidation. As observed in the pure
À
À
ionic liquid system, replacing the [PF
À
6
]
counterion by Cl
and [BF
4
] , in ionic liquids [bmim]Cl and [bmim][BF
4
], led to
2
2
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
New J. Chem., 2008, 32, 283–289 | 285

View Article Online
cyclooctene did not affect the intensity of this peaks, suggest-
ing that the peak could not represent an active species. In pure
acetonitrile solvent and in the presence of [bmim] PW O
3
12 40
and H O , the peak of the original polyoxometalate was
2
2
located at À14.0 ppm, and new small peaks were observed
during reaction at 50 1C (Fig. 3(C)). All these results are in
good accord with those in the activity experiments (Table 1
and Fig. 2). Therefore, it is evident that the ionic liquid
6
[bmim][PF ] could promote the formation of the active
peroxotungstate in the polyoxometalate catalytic system.
It was reported that phosphotungstate has no catalytic
activity for the epoxidation of olefins until the active peroxo-
5
tungstate forms at an appreciable concentration. A induction
period for the formation of epoxides is expected when the
À
3
PW O
40
is used as precursor and if the active peroxotung-
1
2
Fig. 2 Effect of ion salts on epoxidation of cis-cyclooctene with H
catalyzed by [bmim] PW12 40 in acetonitrile at 60 1C for 1 h.
Catalyst] = 2.4 mM, [substrate] = 0.8 M and [C A ]/[Cat.] =
2 2
O
state is formed in a slow rate. In this study, the epoxidation of
3
O
cis-cyclooctene was kinetically studied in the [bmim]3-
+
À
[
+
À
PW12
O40–H
2
O
2
system in order to examine the induction
1000. Ion salts (C A ): A: [bmim][PF ]; B: tetrabutylammonium
6
period. In the ionic liquid [bmim][PF ], epoxidation was
6
bromide; C: cetyltrimethylammonium bromide; D: [bmim]Cl; E:
bmim][BF ].
determined to show first-order kinetics (Fig. 4(A)), and the
À1
[
4
apparent rate constant was evaluated to be 0.060 min . There
2
8
(
Venturello complex). The signals near À12 ppm can be
is no apparent induction period, indicating that the initial
formation of the active peroxo species from the precursor
attributed to polyoxometalates with peroxo groups (usually
2
expressed as PW O ). Further reaction led to the loss of the
8
3 2 2
[bmim] PW12O40 and H O in the ionic liquid is considerably
x
y
signals at À12.0 and À12.1 ppm. We also notice that part of
the species caused the buildup of the resonance at 4.1 ppm as
the other part of the species reconverted to the initial phos-
photungstate (Fig. 3(A), curves c and d). It is implied that the
fast with respect to the epoxidation reaction. By contrast, in
pure acetonitrile, although a very low conversion was ob-
served, the formation of epoxide exhibited an evident induc-
tion period (Fig. 4(B)), which is in agreement with the results
3
0
PW
x
O
y
species are probably intermediates in the formation of
by Alsters and co-workers.
the catalytically active Venturello complex. When cis-cyclooc-
tene (20 mM) was added into the system, the peak at 4.1 ppm
disappeared immediately (Fig. 3(A), curve f), which confirms
that the species with a signal at 4.1 ppm is the active species for
the epoxidation of olefins. When [C H N(CH ) -
During the Ishii–Venturello reaction, the polyoxometalate
3À
precursor (PW12
active epoxidizing species [PO
of excess hydrogen peroxide. Bre
reported that if the H was first mixed with the poly-
O40 in this study) must first transform to the
3
À
in the presence
4
{W(O)(O)
}
2 4
]
2
8
´
geault and co-workers
2 2
O
5
5
2 15
CH ] PW O was treated with H O in acetonitrile at
3
oxometalate in an aqueous solution to form the Venturello
complex before the phase transfer, good catalytic activity was
obtained. In contrast, when the hydrogen peroxide was intro-
duced after the transfer of precursor into organic phase by
3
12 40
2
2
3
À
5
0 1C, besides the signal from the original PW12
À13.8 ppm), a weak peak at 1.3 ppm appeared and became
stronger with reaction time (Fig. 3(B)). The addition of cis-
O
40
(
3
1
Fig. 3 Time-dependent P NMR spectra at 50 1C: (A) [bmim]
acetonitrile–[bmim][PF ] (1 : 1)); (B) [C N(CH 15CH PW12 40 reacted with H
in acetonitrile. Reaction time: (a) 0 min; (b) 15 min; (c) 30 min; (d) 45 min; (e) 90 min. Reaction time after the addition of substrate cis-cyclooctene:
f) 0 min; (g) 10 min.
3
PW12
O
40 reacted with 50 equivalents of H
2
O
2
in a mixed solvent
(
6
5
H
5
2
)
3
]
3
O
2
O
2
in acetonitrile; (C) [bmim] PW12
3
O
40 reacted with H
2 2
O
(
2
86 | New J. Chem., 2008, 32, 283–289
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
lium ionic liquid depends on the three-dimensional network of
anions and cations by weak interaction (such as hydrogen
1
2,31–33
bonds, van der Waals and Coulombic forces).
The
anion identity has an important impact on its properties, such
3
as viscosity, polarity, and H-bond strength. In this study, the
4
inefficiency of epoxidation in [bmim][BF ] (Table 1, entry 9)
4
indicates that the anionic counterion plays an important role
in the formation of the microenvironment. Furthermore, in
the studies on the physicochemistry of ionic liquids, negative
excess molar volumes were always observed when ionic liquids
(
6
such as [bmim][PF ]) were mixed with organic solution (such
as acetonitrile), indicating that the nanostructural organiza-
tion from the ion–ion interactions of ionic liquids does not
have to be broken in the organic solvent, even at very low
3
5–37
concentration of the ionic liquid.
To further reveal the interaction and association between
+
the [bmim] and counterion [PF ] , the electrospray ioniza-
À
6
tion (ESI) spectra were examined. As shown in Fig. 5, the ESI
experiment evidently indicated the presence of the mono-
+
)] and [(bmim)(PF
À
charged species [(bmim)
2
(PF
6
6
)
2
]
in the
acetonitrile solution of the ionic liquid [bmim][PF
6
]. Similar
38
3
results had been reported by Dupont and Zhao. In addi-
2
À
tion, the relative intensity of [(bmim)(PF
6
)
2
]
increased with
6
] (Fig. 6). This
the increase in the concentration of [bmim][PF
change was in accordance with activity of epoxidation in the
acetonitrile solution of the ionic liquid. Accordingly,
[bmim][PF ] when dissolved in acetonitrile can form a special
6
structure, which could enhance the activation of catalyst, and,
consequently, exhibit a good activity of epoxidation.
Fig. 4 Kinetic profiles in [bmim]
cis-cyclooctene with hydrogen peroxide at 60 1C in different solvents:
A) in [bmim][PF ]: (a) the formation of epoxide and (b) the rate of the
3
PW12O40-catalyzed epoxidation of
(
6
epoxide formation vs. reaction time. (B) The formation of epoxide vs.
reaction time in acetonitrile. [Catalyst] = 2.4 mM and [substrate] =
0.8 M.
phase transfer agent, the polyoxometalate exhibited little
catalytic activity. In this case, H O is not accessible to the
2
2
precursor, and the active species is difficult to generate. In our
study, acetonitrile is miscible both with aqueous hydrogen
peroxide, with the substrates and with [bmim]
ever, little active species (Fig. 3(C)) and low epoxidation rate
Table 1 and Fig. 4(B)) were obtained in acetonitrile, indicat-
3
PW12O40. How-
(
ing that the reaction media can greatly affect the formation of
active species and hence significantly change the epoxidation
3
À
reaction rate. In a controlled experiment, [PO
+
with the counterion [bmim] , [Bu
4
{W(O)(O)
} ]
2 4
+
N] or [CH (CH ) -
4
3
2 15
+
3 3
CH ) N] was used directly as catalyst. The epoxidation
(
reaction could occur efficiently in both [bmim][PF ] and pure
6
acetonitrile, indicating that the former process (i.e. the forma-
tion of the active species) is blocked in the pure acetonitrile
environment. Another evidence for the slow formation of
active species in acetonitrile is the existence of an induction
period (Fig. 4(B)). The slow formation of active species also
gives a good interpretation of the low epoxidation rate in other
4 3 2 2
media, such as [bmim][BF ], CH OH and CH Cl (Table 1).
From the above results, it is evident that the some ionic
liquid can provide a special microenvironment for the forma-
tion of the active peroxotungstate in the Ishii–Venturello
reaction. It has been reported that the nature of the imidazo-
6
Fig. 5 ESI-MS spectra of acetonitrile solutions of [bmim][PF ]: (A)
+
the positive ion mode: m/z 139.2 [bmim]
+
and m/z 423.4
À
[bmim–PF –bmim] ; (B) the negative ion mode: m/z 144.9 [PF ]
6
6
À
6 6
and m/z 429.1 [PF –bmim–PF ] .
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
New J. Chem., 2008, 32, 283–289 | 287

View Article Online
2
3
4
5
6
7
8
C. Venturallo and R. D’Aloisio, J. Org. Chem., 1988, 53,
1
Y. Ishii, K. Yamawaki, T. Ura, H. Yamada, T. Yoshida and M.
Ogawa, J. Org. Chem., 1988, 53, 3587.
S. Sakaguchi, Y. Nishiyama and Y. Ishii, J. Org. Chem., 1996, 61,
553.
5
307.
D. C. Duncan, R. C. Chambers, E. Hecht and C. L. Hill, J. Am.
Chem. Soc., 1995, 117, 681.
M. Carraro, L. Sandei, A. Sartorel, G. Scorrano and M. Bonchio,
Org. Lett., 2006, 8, 3671.
Y. Ding, Q. Gao, G. Li, H. Zhang, J. Wang, L. Yan and J. Suo, J.
Mol. Catal. A: Chem., 2004, 218, 161.
´
J.-M. Bregeault, M. Vennat, L. Salles, J.-Y. Piquemal, Y. Mahha,
E. Briot, P. C. Bakala, A. Atlamsani and R. Thouvenot, J. Mol.
Catal. A: Chem., 2006, 250, 177.
9
P. Lei, C. Chen, J. Yang, W. Ma, J. Zhao and L. Zang, Environ.
Sci. Technol., 2005, 39, 8466.
1
1
0 C. Chen, Q. Wang, P. Lei, W. Song, W. Ma and J. Zhao, Environ.
Sci. Technol., 2006, 40, 3965.
1 A. Haimov and R. Neumann, J. Am. Chem. Soc., 2006, 128,
Fig. 6 The effect of the ion pair aggregate of different concentrations
À
of [bmim][PF
2À
6
] in acetonitrile. The target ion pair [PF
6
–bmim–PF
6
]
1
5697.
(CA ) of m/z 429.1 was indicated through ESI-MS spectra in the
12 J. Dupont, R. F. de Souza and P. A. Z. Suarez, Chem. Rev., 2002,
102, 3667.
negative ion mode.
It is also noted that there are some differences in the
1
3 A. A. Valente, %. Petrovski, L. C. Branco, C. A. M. Afonso, M.
Pillinger, A. D. Lopes, C. C. Romao, C. D. Nunes and I. S.
Goncalves, J. Mol. Catal. A: Chem., 2004, 218, 5.
14 M. K. Tse, M. Klawonn, S. Bhor, C. Dobler, G. Anilkumar,
H. Hugl, W. Magerlein and M. Beller, Org. Lett., 2005, 7,
87.
˜
¸
3
chemical shifts of P NMR spectra of the original PW12
1
3À
O
40
¨
¨
(
À14.4 ppm for [bmim]
1 : 1) mixed solvent, À14.0 ppm for [bmim]
N(CH
3
PW12
O
40 in acetonitrile–[bmim][PF
PW12 40 in
15CH PW12
6
]
9
(
3
O
1
1
5 C. E. Song and E. J. Roh, Chem. Commun., 2000, 837.
´
acetonitrile and À13.8 ppm for [C
H
5 5
2
)
3
]
3
O
40
6 A. A. Linden, M. Johansson, N. Hermanns and J.-E. Ba
¨
ckvall, J.
in acetonitrile) (Fig. 3), which confirms that the ionic liquid,
Org. Chem., 2006, 71, 3849.
7 G. D. Allen, M. C. Buzzeo, I. G. Davies, C. Villagran, C. Hardacre
´
and R. G. Compton, J. Phys. Chem. B, 2004, 108, 16322.
8 G. S. Owens and M. M. Abu-Omar, Chem. Commun., 2000,
1165.
9 G. S. Owens, A. Durazo and M. M. Abu-Omar, Chem.–Eur. J.,
2002, 8, 3053.
0 B. S. Chhikara, R. Chandra and V. Tandon, J. Catal., 2005, 230,
1
1
1
2
even mixed with CH CN, can provide a special microenviron-
3
ment for the polyoxometalate species. Therefore, it is reason-
able to propose that the cations and anions of the ionic liquid
could form a microenvironment with nonpolar and polar
3
9–41
regions,
which means that the unique structure is derived
from the anions and the cations arranged in specific dimen-
sions, and here they can interact with the polyoxometalate and
hydrogen peroxide, which can accelerate the activation reac-
tion of polyoxometalate by hydrogen peroxide to generate the
catalytically active species.
4
36.
21 H. Sun, K. Harms and J. Sundermeyer, J. Am. Chem. Soc., 2004,
26, 9550.
2 V. Cimpeanu, A. N. Pa
1
2
ˆ rvulescu, V. I. Paˆ rvulescu, D. T. On, S.
Kaliaguine, J. M. Thompson and C. Hardacre, J. Catal., 2005, 232,
60.
2
3 K. Yamaguchi, C. Yoshida, S. Uchida and N. Mizuno, J. Am.
Chem. Soc., 2005, 127, 530.
Conclusions
24 J. Dupont, C. S. Consorti, P. A. Z. Suarez and R. F. de Souza,
Org. Synth., 2000, 79, 236.
Ionic liquids could not only be used as solvent for the reuse of
catalyst, but also be beneficial to epoxidation of olefins
25 L. Cammarata, S. G. Kazarian, P. A. Salter and T. Welton, Phys.
Chem. Chem. Phys., 2001, 3, 5192.
2
6 M. H. Chiang, J. A. Dzielawa, M. L. Dietz and M. R. Antonio, J.
Electroanal. Chem., 2004, 567, 77.
3 2 2
catalyzed by [bmim] PW12O40 and H O . In the ionic liquid,
the special microenvironment was supplied by the interactions
of cations and anions, in which the generation of the active
peroxotungstate from [bmim] PW O and H O could be
27 Hydrogen peroxide was analyzed by the iodometric method.
28 C. Aubry, G. Chottard, N. Platzer, J.-M. Bregeault, R. Thouvenot,
F. Chauveau, C. Huet and H. Ledon, Inorg. Chem., 1991, 30,
409.
9 L. Salles, C. Aubry, R. Thouvenot, F. Robert, C. Dore
Morin, G. Chottard, H. Ledon, Y. Jeannin and J.-M. Bre
Inorg. Chem., 1994, 33, 871.
0 P. T. Witte, P. L. Alsters, W. Jary, R. Mu
´
3
12 40
2
2
4
accelerated significantly. In some sense, the ionic liquid
bmim][PF ] could be considered as a cocatalyst to promote
2
3
´
mieux-
´
geault,
[
6
the formation of active peroxotungstate from phosphotung-
state and hydrogen peroxide.
¨
llner, P. Po
¨
chlauer, D.
Sloboda-Rozner and R. Neumann, Org. Process Res. Dev., 2004, 8,
24.
5
3
1 D. Xiao, J. R. Rajian, A. Cady, S. Li, R. A. Bartsch and E. L.
Quitevis, J. Phys. Chem. B, 2007, 111, 4669.
Acknowledgements
3
2 F. C. Gozzo, L. S. Santos, R. Augusti, C. S. Consorti, J. Dupont
and M. N. Eberlin, Chem.–Eur. J., 2004, 10, 6187.
The generous financial support by 973 project
(
2007CB613306), NSFC (No. 20537010, 20520120221, and
0772129), and CAS is gratefully acknowledged.
33 C. S. Consorti, P. A. Z. Suarez, R. F. de Souza, R. A. Burrow,
D. H. Farrar, A. J. Lough, W. Loh, L. H. M. da Silva and
J. Dupont, J. Phys. Chem. B, 2005, 109, 4341.
2
3
4 J. G. Huddleston, A. E. Visser, W. M. Reichert, H. D.
Willauer, G. A. Broker and R. D. Rogers, Green Chem., 2001, 3,
References
1
56.
1
C. Venturello, R. D’Aloisio, J. C. J. Bart and M. Ricci, J. Mol.
Catal., 1985, 32, 107.
35 M. T. Zafarani-Moattar and H. Shekaari, J. Chem. Eng. Data,
2005, 50, 1694.
2
88 | New J. Chem., 2008, 32, 283–289
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
3
6 J. Wang, A. Zhu, Y. Zhao and K. Zhuo, J. Solution Chem., 2005,
4, 585.
7 G. Hong, J. Jacquemin, P. Husson, M. F. Costa Gomes, M.
Deetlefs, M. Nieuwenhuyzen, O. Sheppard and C. Hardacre, Ind.
Eng. Chem. Res., 2006, 45, 8180.
39 U. Schro
¨
der, J. D. Wadhawan, R. G. Compton, F. Marken, P. A.
3
Z. Suarez, C. S. Consorti, R. F. de Souza and J. Dupont, New J.
Chem., 2000, 24, 1009.
40 J. Dupont, Braz. Chem. Soc., 2004, 15, 341.
41 J. N. A. Canongia Lopes and A. A. H. Pa
´
2006, 110, 3330.
3
dua, J. Phys. Chem. B,
38 P. J. Dyson, J. S. McIndoe and D. Zhao, Chem. Commun., 2003, 508.
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
New J. Chem., 2008, 32, 283–289 | 289