Methyltrioxorhenium-catalyzed epoxidation of olefins with hydrogen peroxide as an oxidant and pyridine N-oxide ionic liquids as additives

Article abstract of DOI:10.1016/j.molcata.2012.09.017

Four ionic liquids (ILs) with both a pyridine N-oxide moiety and an imidazolium moiety combined by an amide spacer were synthesized through a series of reactions including amidation, peroxidation, quaterization and anion exchange reaction. Their structures were fully characterized by ¹H NMR, FT-IR, UV-vis and HR-MS. The ionic liquids were used respectively as additives in the methyltrioxorhenium (MTO) catalyzed epoxidation of olefins with 30% H₂O₂ as an oxidant. The catalytic results displayed that the ILs had excellent performances in suppressing epoxide ring-opening reaction, which led to the significant improvement of the selectivity of the MTO-catalyzed epoxidation with low loadings compared to other substances as additives. The coexistence of the pyridine N-oxide and imidazolium moieties in the structures of ILs is necessary in improving the MTO-catalyzed epoxidation reaction. It was also displayed that the improvement degree on the selectivity of epoxidation depended on the type of anion of the ILs, but not the position of the substituent with imidazolium moiety in the ring of pyridine N-oxide. Meanwhile, the results also showed that the introduction of the ILs caused the decrease of the epoxidation rate, but this side effect was small compared to those of other substances used as additives.

Full text of DOI:10.1016/j.molcata.2012.09.017

Journal of Molecular Catalysis A: Chemical 366 (2013) 149–155
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Methyltrioxorhenium-catalyzed epoxidation of olefins with hydrogen peroxide
as an oxidant and pyridine N-oxide ionic liquids as additives
Yuecheng Zhang, Zhaozhao Li, Xiaohui Cao, Jiquan Zhao∗
School of Chemical Engineering and Technology, Hebei University of Technology, Tianjin 300130, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 May 2012
Received in revised form
10 September 2012
Accepted 16 September 2012
Available online 25 September 2012
Four ionic liquids (ILs) with both a pyridine N-oxide moiety and an imidazolium moiety combined by an
amide spacer were synthesized through a series of reactions including amidation, peroxidation, quateri-
zation and anion exchange reaction. Their structures were fully characterized by 1H NMR, FT-IR, UV–vis
and HR–MS. The ionic liquids were used respectively as additives in the methyltrioxorhenium (MTO)
catalyzed epoxidation of olefins with 30% H2O2 as an oxidant. The catalytic results displayed that the ILs
had excellent performances in suppressing epoxide ring-opening reaction, which led to the significant
improvement of the selectivity of the MTO-catalyzed epoxidation with low loadings compared to other
substances as additives. The coexistence of the pyridine N-oxide and imidazolium moieties in the struc-
tures of ILs is necessary in improving the MTO-catalyzed epoxidation reaction. It was also displayed that
the improvement degree on the selectivity of epoxidation depended on the type of anion of the ILs, but
not the position of the substituent with imidazolium moiety in the ring of pyridine N-oxide. Meanwhile,
the results also showed that the introduction of the ILs caused the decrease of the epoxidation rate, but
this side effect was small compared to those of other substances used as additives.
Keywords:
Methyltrioxorhenium
Pyridine N-oxide
Ionic liquids
Olefins
Epoxidation
Hydrogen peroxide
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
synthesized to evaluate their improvement on the MTO-catalyzed
epoxidation [14–37]. We also synthesized some Schiff-bases as lig-
Epoxides are valuable intermediates in organic chemical, fine
chemical and pharmaceutical synthesis. The epoxidation of olefins
is one of the most efficient strategies for obtaining epoxides [1].
Among all of the procedures the epoxidation of olefins with hydro-
gen peroxide (H2O2) as an oxidant has received special attention
due to its high atom efficiency and environment friendliness [2–5].
However, the oxidation activity of H2O2 is so weak that gener-
ally appropriate catalysts such as transition metal complexes are
needed to activate it [3,6,7]. In 1991, Herrmann and co-workers
[8,9] found that methyltrioxorhenium(VII) (CH3ReO3, MTO) as a
catalyst was very powerful in the epoxidation of olefins with H2O2
as an oxidant. However, the MTO/H2O2 catalytic system is not per-
fect due to the fact that the strong Lewis acidity of central rhenium
atom (Re) can promote the ring opening of epoxides leading to the
formation of diols in the presence of water during the reaction,
meanwhile, MTO and the active species derived from MTO decom-
pose in the aqueous solutions [10–13]. When trying to overcome
both the problems it was found that the addition of Lewis base
ligands, containing nitrogen and (or) oxygen donor atoms, could
reduce the Lewis acidity of MTO and increase the selectivity of the
epoxides. Therefore, a wide range of ligands has been chosen or
ands for the MTO-catalyzed epoxidation of olefins with 30% H2O2
as an oxidant, which showed excellent selectivity for the forma-
tion of epoxides with moderate or poor catalytic activity [38–41].
All previous results have shown that pyridine and its derivatives
are among the most efficient additives at increasing the selectiv-
ity of the reaction. However, generally significant excess of the
Lewis base additives is necessary to achieve excellent catalytic per-
formance in such cases [17,21,27], which is unfavorable from the
viewpoints of cost and separation. Besides, if the additives are too
basic, then, MTO will decompose to catalytically inert perrhenic
acid and methanol [11,25,26,32]. Much efforts have been made to
seek additives that can suppress the ring opening side reaction
in low loading, while not promote the decomposition of MTO or
its derivatives. It was found that some oxygen donor ligands can
coordinate with MTO to form stable complexes [19,31,33–37]; the
complexes pre-prepared or generated in situ not only suppressed
the ring opening side reaction, but also showed long lifetime com-
pared to the cases with nitrogen donor ligands as additives in
epoxidation of olefins. Moreover, in contrast to N-donor adducts,
no pronounced ligand excess is necessary to achieve high yields
and selectivity in olefin epoxidation catalysis.
On the other hand, MTO-catalyzed epoxidations carried out in
ILs were reported in literature [42–44]. Several olefinic substrates
with different structure characters were oxidized at room temper-
ature (RT) to the corresponding epoxides with fair to excellent
∗
Corresponding author. Tel.: +86 22 60202926/4279; fax: +86 22 60202926.
E-mail address: zhaojq@hebut.edu.cn (J. Zhao).
1381-1169/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2012.09.017

150
Y. Zhang et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 149–155
Scheme 1. Synthesis of pyridine N-oxide ionic liquids.
yields. In most cases, the epoxidation rates are at least compa-
rable, if not higher than those reported previously. For example,
the epoxidation of cyclooctene in four different water-equilibrated
ILs, [BMIM]NTf2, [BMIM]PF6, [BMIM]BF4, and [C8MIM]PF6, with
H2O2 as an oxidant and Schiff- and Lewis-base adducts of MTO
acting as active catalytic species gave higher yields compared to
the solvent-free systems [43]. Recently, we found that 1-methyl-
3-(butyl-4-sulfonate) imidazoliumbetaine (MBSIB), a zwitter ionic
compound similar to ILs in structure, showed good performances
as an additive in enhancing the MTO-catalyzed epoxidation of
alkenes with 30% H2O2 as an oxidant [45]. The introduction of 10-
fold excess MBSIB significantly improved the selectivity of epoxide
without reducing the reaction rate. Furthermore, the MTO–MBSIB
system was stable throughout the entire catalytic run and increased
the TON of the epoxidation of alkenes. Our work in combination
with the literature described above encourages us to design some
new type ligands with both oxygen donor and ionic liquid moieties
as shown in Scheme 1. As expected, the catalytic results showed
that low loadings of the ligands can suppress the ring opening of
epoxides and extend the lifetime of MTO in the MTO-catalyzed
epoxidation of olefins with 30% H2O2 as an oxidant. Besides, the
high stability and solubility of the ligands in water make them easy
separation with the organics and recycling. Herein, we reported the
results.
2. Experimental
2.1. Materials and reagents
2-Aminomethyl pyridine and methyltrioxorhenium (MTO)
were purchased from Alfa Aesar. Cyclohexene (99%), 1-octene
(99%) and styrene (99%) were purchased from Acros. Pyridine N-
oxide (PyNO) was prepared according to a previously reported
method [46]. 1-n-Butyl-3-methylimidazolium hexafluorophos-
phate ([bmim]PF6) was prepared as described in literature [47].
4-Aminomethyl pyridine was prepared by us in high purity. All of
the other reagents and solvents were obtained from commercial
sources and used as received without further purification.
2.2. Physical measurements
The 1H NMR spectra were recorded on a Bruker AC-P400 instru-
ment using CDCl3, DMSO or D2O as solvent and TMS as internal

Y. Zhang et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 149–155
151
standard. IR spectra were recorded on a Bruker Vector-22 spec-
trophotometer using KBr pellets as the IR matrix. The UV–vis
spectra were recorded on a Varian Cary 300 spectrophotometer.
Melting points were determined on a Perkin XT-4 microscopic
analyzer. ESI mass spectra were recorded on a LCQ Advanced high-
resolution mass spectrometer.
ethanol was stirred at 75 ◦C for 5 h under nitrogen atmosphere.
After evaporation of ethanol, acetonitrile was added to the mix-
ture causing precipitation of C1 as a white solid. This solid was
recovered by filtration, washed twice with acetonitrile and dried
in vacuum. Finally the product was obtained in 89% yield. m.p.
210.5–212.5 ◦C; 1H NMR (D2O) ı: 3.82(s, 3H, NCH3), 4.56(s, 2H,
COCH2), 5.08(s, 2H, PyCH2NHCO), 7.37(s, 2H, IMI.H), 7.42–7.50(m,
2H, Py.H), 7.61(d, J = 8.0 Hz, 1H, Py.H), 8.26(d, J = 6.8 Hz, 1H, Py.H),
8.69(s, 1H, IMI.H) ppm; IR (KBr) ␯: 3424, 1686, 1567, 1493, 1433,
1265, 1221, 1177, 1111, 1031 cm−1; UV–vis (CH3OH) ꢁmax: 214,
262 nm; MS(ESI) m/z: 247.1190 [M+−HCl].
2.3. Preparation of the pyridine N-oxide ionic liquids
2.3.1. Synthesis of A1 and A2
2-Chloro-N-(pyridin-2-ylmethyl)acetamide (A1). To a suspen-
sion of 2-aminomethyl pyridine (6.0 g, 0.056 mol) and K2CO3
(4.0 g, 0.028 mol) in dichloromethane (50 mL) was slowly added 2-
chloroacetyl chloride (3.3 g, 0.028 mol) in dichloromethane (20 mL)
at 0 ◦C under nitrogen atmosphere. The reaction mixture was
stirred at room temperature for 1 h, then the mixture was fil-
tered and the filtrate was evaporated under reduced pressure. The
residue was purified by flash column chromatography on silica
gel using a mixture of EtOAc/petroleum ether (30–60 ◦C) 1:1 as
eluent to afford product A1 as a yellowish liquid (90%). 1H NMR
(CDCl3) ı: 4.13(s, 2H, COCH2Cl), 4.61(d, J = 5.2 Hz, 2H, PyCH2NHCO),
7.21–7.28(m, 2H, Py.H), 7.66(d, J = 1.6 Hz, 1H, Py.H), 7.85(s, 1H, NH),
8.58(d, J = 4.8 Hz, 1H, Py.H); IR (KBr) ꢀ: 3430, 3069, 1694, 1632, 1562,
1517, 1461, 1362, 1259, 1101, 1026, 777 cm−1; UV–vis (CHCl3)
ꢁmax: 267 nm.
2-Chloro-N-(pyridin-4-ylmethyl)acetamide (A2). The reaction
procedure of preparing A2 is similar to that of A1. The residue
was purified by flash column chromatography on silica gel using
a mixture of EtOAc/petroleum ether (30–60 ◦C) 1:1 mixed with
appropriate amount of acetic acid as eluent. The eluent was evapo-
rated in vacuum to give A2 acetate as a yellowish liquid. The product
was used for the next step directly.
1-Methyl-3-(2-oxo-2-(N-(N-oxide-pyridin-4-ylmethyl))-
ylamine)ethyl-imidazolium chloride (C2). C2 was prepared similar
to C1. Yield 75%; m.p. 105.0–107.5 ◦C; 1H NMR (D2O, 400 MHz) ı:
3.71(s, 3H, NCH3), 4.33(s, 2H, COCH2), 4.97(s, 2H, PyCH2NHCO),
7.26–7.27(m, 2H, IMI.H), 7.31(d, J = 6.8 Hz, 2H, Py.H), 8.05–8.07(t,
J = 4.8 Hz, 2H, Py.H), 8.60(s, 1H, IMI.H) ppm; IR (KBr) ꢀ: 3424, 1696,
1563, 1495, 1432, 1270, 1227, 1175, 1109, 1029 cm−1; UV–vis
(CH3OH) ꢁmax: 214, 262 nm; MS(ESI) m/z: 247.1189 [M+−HCl].
2.3.4. Synthesis of D1 and D2
1-Methyl-3-(2-oxo-2-(N-(N-oxide-pyridin-2-ylmethyl))-
ylamine)ethyl-imidazolium hexafluorophosphate (D1). To a solution
of C1 (13.5 g, 0.012 mol) in water (25 mL) was added potassium
hexafluorophosphate (4.7 g, 0.024 mmol), and the mixture was
stirred at 50 ◦C for 30 h. After removal of water in vacuum,
the residue was dissolved in acetonitrile and the insoluble salt
precipitate was removed by filtration. Finally, the filtrate was
concentrated in vacuum to give D1 as a yellowish solid in a
yield of 95%. m.p. 172.0–174.5 ◦C; 1H NMR (D2O) ı: 3.86(s, 3H,
NCH3), 4.60(s, 2H, COCH2), 5.11(s, 2H, PyCH2NHCO), 7.41(s, 2H,
IMI.H), 7.48–7.53(m, 2H, Py.H), 7.63 (d, J = 7.6 Hz, 1H, Py.H), 8.29(d,
J = 6.4 Hz, 1H, Py.H), 8.72(s, 1H, IMI.H) ppm; IR (KBr) ꢀ:3425, 1698,
1562, 1492, 1438, 1369, 1232, 1179, 1113, 1037, 843 cm−1; UV–vis
(CH3OH) ꢁmax: 214, 262 nm.
2.3.2. Synthesis of B1 and B2
2-Chloro-N-(N-oxide-pyridin-2-ylmethyl)acetamide (B1). A solu-
tion of A1 (13.3 g, 0.018 mol) and urea hydrogen peroxide (UHP)
(2.0 g, 0.011 mol) in acetic acid (20 mL) was stirred at 75 ◦C for 3 h.
Then additional UHP (4.0 g, 0.022 mol) was added to this solution
and the mixture was stirred at 75 ◦C for further 7 h. After evapo-
ration of the solvent, the resulting residue was neutralized with
saturated aqueous K2CO3, and then extracted with chloroform.
The combined organic layers were dried over anhydrous Na2SO4
and concentrated to give yellowish oil. The oil was recrystallized
from ethanol–cyclohexane to give B1 as a white solid (70%): m.p.
119.5–120.5 ◦C. 1H NMR (CDCl3) ı: 4.0(s, 2H, COCH2Cl), 4.68(d,
J = 6.0 Hz, 2H, PyCH2NHCO), 7.26–7.29(m, 2H, Py.H), 7.45–7.49(t,
J = 4.8 Hz, 1H, Py.H), 8.05(s, 1H, NH), 8.25–8.27(m, 1H, Py.H) ppm;
IR (KBr) ꢀ: 3430, 1663, 1543, 1440, 1432, 1271, 1212, 1112, 1027,
761 cm−1; UV–vis (CHCl3) ꢁmax: 271 nm.
2-Chloro-N-(N-oxide-pyridin-4-ylmethyl)acetamide (B2). The
reaction procedure of preparing B2 is similar to that of B1. The
reaction mixture was evaporated under vacuum to remove the
solvent. The resulting residue was dissolved in methanol and the
solution was neutralized with solid Na2CO3. Then the mixture was
filtered and the filtrate was evaporated under reduced pressure.
The residue was purified by flash column chromatography on
silica gel using a mixture of EtOAc/MeOH 15:1 as eluent to give
product B2 as a yellowish liquid after removing the solvent in
vacuum. 1H NMR (DMSO) ı: 4.13(s, 2H, COCH2Cl), 4.25(d, J = 6.0 Hz,
2H, PyCH2NHCO), 7.22–7.26(m, 2H, Py.H), 8.11–8.15(m, 2H, Py.H),
8.91(s, 1H, NH) ppm.
1-Methyl-3-(2-oxo-2-(N-(N-oxide-pyridin-4-ylmethyl))-
ylamine)ethyl-imidazolium hexafluorophosphate (D2). D2 was
prepared similar to D1. Yield 75%; m.p. 85.0–87.5 ◦C; 1H NMR
(D2O, 400 MHz) ı: 3.88(s, 3H, NCH3), 4.63(s, 2H, COCH2), 5.14(s,
2H, PyCH2NHCO), 7.43(s, 2H, IMI.H), 7.50–7.56(m, 2H, Py.H), 7.65
(d, J = 6.4 Hz, 1H, Py.H), 8.32(d, J = 4.4 Hz, 1H, Py.H), 8.75(s, 1H,
IMI.H) ppm; IR (KBr) ꢀ: 3425, 1693, 1558, 1497, 1433, 1273, 1221,
1111, 1029, 842 cm−1; UV–vis (CH3OH) ꢁmax: 214, 262 nm.
2.4. Catalytic reaction
The epoxidation reactions were performed in a 25 mL round
bottomed flask immersed in a temperature-controlled water bath.
In a typical experiment, 5 mmol of substrate, 7 mL of methanol
and 0.05 mmol of MTO were mixed in this flask. Subsequently,
aqueous H2O2 (30 wt.%, 10 mmol) was added to initiate the reac-
tion. Samples were collected at regular intervals and analyzed
by a Shandong Lunan Ruihong Gas Chromatograph (SP-6800A)
equipped with a 30 m × 0.25 mm SE 30 capillary column and an FID
detector.
2.5. Lifetime test
Once the previous substrate was exhausted, fresh substrate and
aqueous H2O2 were sequentially added into the reaction mixture,
and the reaction was run again and monitored by GC.
2.3.3. Synthesis of C1 and C2
2.6. Recovery of C1
1-Methyl-3-(2-oxo-2-(N-(N-oxide-pyridin-2-ylmethyl))-
ylamine)ethyl-imidazolium chloride (C1). A solution of B1 (13 g,
0.015 mol) and 1-methylimidazole (1.9 g, 0.023 mol) in absolute
After the first run, the reaction mixture was evaporated
under high vacuum at 40 ◦C to remove the methanol, water,

152
Y. Zhang et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 149–155
cyclohexene oxide and trace of cyclohexanediol formed in the
catalytic run. The residue was analyzed by GC to make sure no
presence of cyclohexene oxide and cyclohexanediol. Then the
residue was subjected to the next catalytic run as described in
Section 2.4.
3.2. Test of ILs as additives in the MTO-catalyzed epoxidation
3.2.1. The epoxidation of different olefins catalyzed by
MTO/pyridine N-oxide ionic liquids
In the beginning, methanol, methylene chloride and acetonitrile
were chosen as solvents to investigate the influence of the four
ILs on the MTO-catalyzed epoxidation with 30% H2O2 as an oxi-
dant. The results showed that the epoxidation could not undergo
smoothly in methylene chloride and acetonitrile due to the insol-
ubility of the ILs in the solvents. Therefore, all the epoxidation
reactions were run in methanol. Cyclohexene, styrene and 1-octene
were chosen as the substrates with different structure characters
to evaluate the effect of the four ILs as additives on the MTO-
catalyzed epoxidation of olefins with 30% H2O2 as an oxidant at
0 ◦C and 20 ◦C, respectively, and the results are summarized in
Table 1. It can be seen that all the substrates underwent epoxidation
reaction under the conditions employed, however, the epoxidation
rate are different. Cyclohexene is epoxidized most easily, followed
by styrene, and 1-octene is epoxidized with the most difficulty.
Similar phenomenon was also observed in the cases with other
substances as additives [17,32,38–42]. Herein, the conversion of
cyclohexene reached higher than 97% in all cases when the reac-
tion was run at 0 ◦C for 6 h (entries 2–5, Table 1); however, the
conversions of styrene and 1-octene were only about 47% and
27%, respectively, even if the reaction was run for 12 h (entries
12–15, 23–26, Table 1). The results in Table 1 displayed that the
four ILs as additives dramatically increased the selectivity of the
epoxidation reaction for all the substrates especially at higher tem-
perature, however, decreased the reaction rate in some degree
compared to the neat MTO-catalyzed epoxidation, which were also
3. Results and discussion
3.1. Preparation and characterization of C1, C2, D1 and D2
Herein, the ionic liquids (ILs, C1, C2, D1 and D2) with both
a pyridine N-oxide moiety and an imidazolium moiety were
prepared as shown in Scheme 1. The two moieties were combined
with an amide spacer (–CH2CONH–). First, 2(or 4)-aminomethyl
pyridine reacted with 2-chloroacetyl chloride in dichloromethane
in the presence of potassium carbonate to afford 2-chloro-N-(N-
oxide-pyridin-2(or 4)-ylmethyl)acetamide A1 (or A2) smoothly.
A1 or A2 was oxidized with UHP in acetic acid to give 2-chloro-
N-(N-oxide-pyridin-2(or 4)-ylmethyl)acetamide B1 or B2 in
good yield. Respective quaternization of 1-methylimidazole
with B1 and B2 afforded 1-methyl-3-(2-oxo-2-(N-(N-oxide-
pyridin-2-ylmethyl))-ylamine)ethyl-imidazolium chloride (C1)
and
1-methyl-3-(2-oxo-2-(N-(N-oxide-pyridin-4-ylmethyl))-
ylamine)ethyl-imidazolium chloride (C2) in excellent and
good yields. Finally, the ILs D1 and D2 were obtained in high
yields by anion exchange reaction to replace chloric anion
for hexafluorophosphate. The ILs C1, C2, D1 and D2 were
characterized by 1H NMR, FT-IR, UV–vis and high resolved
MS.
Table 1
The epoxidation of different olefins catalyzed by MTO/pyridine N-oxide ionic liquids.
Entry
Substrate
Temperature (◦C)
Catalyst
Time (h)
Conversion (%)
Selectivity (%)
1
2
3
4
5
6
7
8
9
0
0
0
0
0
20
20
20
20
20
MTO
6
6
6
6
99
97
98
97
97
99
92
93
92
93
70
99
99
99
99
27
95
90
95
90
Cyclohexene
MTO/C1
MTO/D1
MTO/C2
MTO/D2
MTO
MTO/C1
MTO/D1
MTO/C2
MTO/D2
6
1.5
1.5
1.5
1.5
1.5
10
11
12
13
14
15
16
17
18
19
20
21
Styrene
0
0
0
0
0
20
20
20
20
20
20
MTO
12
12
12
12
12
3
3
3
3
3
60
46
47
47
47
61
44
48
44
49
84
13
85
81
86
82
6
84
69
86
72
99
MTO/C1
MTO/D1
MTO/C2
MTO/D2
MTO
MTO/C1
MTO/D1
MTO/C2
MTO/D2
MTO/C1a
3
22
23
24
25
26
27
28
29
30
31
32
1-Octene
0
0
0
0
0
20
20
20
20
20
20
MTO
12
12
12
12
12
12
12
12
12
12
12
30
26
27
26
26
85
48
67
48
68
86
85
99
99
99
99
78
98
87
99
88
99
MTO/C1
MTO/D1
MTO/C2
MTO/D2
MTO
MTO/C1
MTO/D1
MTO/C2
MTO/D2
MTO/C1a
Reaction conditions: olefin 5 mmol, 30% H2O2 10 mmol, MTO 0.05 mmol, ligand 0.05 mmol, methanol 7 mL.
a
MTO 0.10 mmol; C1 0.10 mmol.

Y. Zhang et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 149–155
153
Table 2
observed in other basic additives promoted MTO-catalyzed epoxi-
The effect of the molar ratio of C1 to MTO on the epoxidation of cyclohexene.
dation [27,40].
It is noteworthy that the effects of ILs on the epoxidation
depended on the type of anion in some degree. The ILs with Cl− as
anion performed better in improving the selectivity of the epoxides
than the ones with PF6− as anion. And the higher the reaction tem-
perature was, the more obvious difference between the two anions
was observed (entries 7–10, 17–20, 28–31, Table 1). For example,
when the epoxidation of cyclohexene was run at 20 ◦C for 1.5 h with
C1 and D1 or C2 and D2 as additives, respectively, the magnitudes
of the selectivity of the epoxide were about 95% and 90%, respec-
tively (entries 7–10, Table 1). The reason of D1 and D2 inferior to C1
and C2 in improving epoxidation selectivity may be the instability
Entry
C1/MTO
(molar ratio)
Time (h)
Conversion (%)
Selectivity (%)
1
2
3
4
0/1
0.5/1
1/1
1.5
2
3
99
99
99
75
27
86
93
95
2/1
3
Reaction conditions: cyclohexene 5 mmol, 30% H2O2 10 mmol, MTO 0.05 mmol,
methanol 7 mL, reaction temperature 20 ◦C.
the amount of epoxide to 1,2-diols owing to ring-opening of
epoxides.
−
of PF6 anion toward hydrolysis in contact with moisture, which
has been noted in a number of reports [49–51]. However, the posi-
tion of the substitute with imidazolium moiety in the pyridyl ring
seems not important, because no difference between C1 and C2
or D1 and D2 in effecting the epoxidation reaction was observed
throughout the reactions of all the substrates at different temper-
ature. For example, when the epoxidation of cyclohexene was run
at 20 ◦C for 1.5 h, the selectivity of the epoxide was around 95%, the
conversion of cyclohexene was 92% in both cases of C1 and C2 as
additives, respectively (entries 7, 9, Table 1); and in the same reac-
tion conditions the selectivity of the epoxide and the conversion of
cyclohexene were 90% and 93% in both cases of D1 and D2 as addi-
tives (entries 8, 10, Table 1), respectively. From the structures of the
ILs, it is expected that only N-oxide group as an electron donor can
coordinate with the center Re of MTO to generate the complexes
in situ as shown in Fig. 1. No difference between C1 and C2 or D1
and D2 as additives in effecting the epoxidation reaction indicated
that the amide spacer between the pyridyl N-oxide and the imid-
azolium moiety in both the structures is long enough to avoid the
steric hindrance of imidazolium moiety in the epoxidation reaction.
Moreover, the influence of the loading amount of MTO and C1
(MTO:C1 = 1:1) on the epoxidation of styrene and 1-octene was also
investigated. As show in Table 1, with the increasing of MTO and
C1 loadings from 1 mol% to 2 mol%, both the conversion of olefins
and selectivity of the epoxides were improved (entries 17, 21, 28,
32, Table 1). When the epoxidation of styrene catalyzed by 2 mol
% MTO/C1 was run at 20 ◦C for 3 h, the conversion of styrene and
the selectivity of epoxide were 84% and 99% corresponding to 44%
and 84% in case of 1% of MTO/C1 loading (entries 17, 21, Table 1);
when the epoxidation of 1-octene was run at 20 ◦C for 12 h, the
conversion of 1-octene increased from 48% to 86%, meanwhile, the
selectivity of epoxide increased from 98% to 99% with the increas-
ing of MTO/C1 from 1% to 2% mol (entries 28, 32, Table 1). This
results showed that increasing the loading amount of catalyst accel-
erated the epoxidation reaction, meantime, decreased the exposure
time of the generated epoxides with water, therefore, decreased
3.2.2. Effect of C1 to MTO molar ratio on the epoxidation of olefins
Because the four ILs had similar performances in improvement
of the MTO-catalyzed epoxidation, C1 was used as a representative
to evaluate the effect of IL to MTO molar ratio on the epoxidation
of olefins and the results are given in Table 2. It was found that
the selectivity of the epoxide increased with the increasing of C1
to MTO molar ratio. The observed results seem consistent with the
view that the enhanced selectivity was due to the in situ coordi-
nation of the pyridine N-oxide moiety of C1 to MTO (as shown in
Fig. 1), which decreased the acidity of the Re, thereby inhibiting
epoxide ring-opening. Accordingly, with the increasing of C1 the
equilibrium shifted toward the MTO–C1 adduct, which promoted
the concentration of adduct in solution and decreases the acidity of
MTO, so that the selectivity of the epoxide increased. Meanwhile, it
was also noted that only 2 percentage point increase of the selec-
tivity of epoxide was received with the increasing of C1 to MTO
molar ratio from 1:1 to 2:1, indicating that low concentration of C1
was enough to suppress the epoxide ring opening (entries 3 and 4,
Table 2). The observed results indicated that C1 is different from
those ones that generally high molar ratio of additive to MTO was
necessary to suppress the side reaction [17,18].
3.2.3. Comparison of different ligands on the epoxidation of
cyclohexene
The results displayed that low loading of C1 can suppress the
epoxide ring opening, meanwhile, not slow down the epoxidation
reaction significantly compared to other additives reported in the
literature [10,42,48]. For understanding the mechanism leading to
the excellent performance of C1 as an additive in the MTO-catalyzed
epoxidation, some other substances including [bmim]PF6, pyridine
N-oxide and B1 were respectively used as additives in the reaction,
their effects on the reaction were examined and compared with
that of C1 by keeping the epoxidation at the same progress (99%
conversion). It can be seen from Table 3 that when the molar ratio
of [bmin]PF6 to MTO was 1:1 in the MTO-catalyzed epoxidation of
O
O
N
N
N
N
HN
HN
O
X-
X-
N
N
O
CH3
CH3
O
O
Re
Re
O
O
O
O
X=Cl , PF6
X=Cl , PF6
Fig. 1. Coordination of ILs with MTO.

154
Y. Zhang et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 149–155
Table 3
100
Effects of different ligands on the epoxidation of cyclohexene.
Entry
Additive
Time (h)
Conversion (%)
Selectivity (%)
80
60
40
20
0
1
2
3
4
5
6
7
8
None
6
6
5
7.5
7.5
8
8
8
>99
>99
>99
>99
>99
>99
97
70
68
45
84
85
92
>99
>99
(1)
(2)
(3)
(4)
[bmim]PF6
[bmim]PF6
PyNO
a
B1
Pyridine
Pyridineb
C1
>99
Reaction conditions: cyclohexene 5 mmol, 30% H2O2 10 mmol, MTO 0.05 mmol,
ligand 0.05 mmol, methanol 7 mL, reaction temperature 0 ◦C.
a
[bmim]PF6 0.5 mmol.
Pyridine 0.6 mmol.
b
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Time(h)
cyclohexene, [bmin]PF6 almost had no effect, neither on the rate of
the reaction nor the selectivity of epoxide (entries 1 and 2, Table 3).
With the molar ratio of [bmin]PF6 to MTO increasing to 10:1, the
conversion of cyclohexene reached 99% within 5 h, however, the
selectivity decreased to 45% (entry 3, Table 3). It is obvious that
large excess of [bmin]PF6 accelerated both the epoxidation and the
epoxide ring-opening rate at the same time. In cases of PyNO and
B1 as additives, both of which have the pyridine N-oxide moiety as
that in C1, the selectivity of epoxide was around 85%, respectively,
when the conversion of cyclohexene reached 99% (entries 4 and 5,
Table 3). And longer time (7.5 h) was needed to finish the epoxi-
dation reaction similar to that with 3-cyanopyridine N-oxide as an
additive in literature [25]. In case of pyridine as an additive, as high
as 12:1 of molar ratio of pyridine to MTO was needed to suppress
the epoxide ring-opening reaction, however, the epoxidation reac-
tion was slowed down compared to the case of C1 as an additive
(entries 7, 8, Table 3). Based on the above results, we can conclude
that for the simple structural additives the coordination capabil-
ity of the additive to MTO plays an central role in suppressing the
epoxide ring-opening reaction as reported in literature [20,40], the
stronger the coordination capability is, the higher the selectivity
of epoxide is obtained. However, when we consider C1, it is not
appropriate just attributing the enhanced selectivity of epoxide to
the in situ coordination of the pyridine N-oxide moiety of C1 to
MTO, because the coordination capacity of pyridine is stronger than
that of C1 in principle. Moreover, PyNO and B1 just lack the imid-
azolium moiety compared to C1 from the structural point of view;
however, their improvement on the epoxidation reaction is infe-
rior to that of C1. Therefore, the coexistence of pyridine N-oxide
and quaternary ammonium in one structure is necessary to greatly
enhance the capacity to suppress the epoxide ring-opening, with-
out slowing down the epoxidation reaction largely. It is difficult to
clarify what causes this result. In the beginning, we attributed this
to the simultaneous binding of pyridine N-oxide and quaternary
ammonium with MTO to stabilize MTO and reduce the acidity of
the center Re atom of MTO. DFT calculation results did not support
this assumption. The possible reason is the salvation of the quater-
nary ammonium moiety [52], which enhanced the solubility and
stabilize the complexes as shown in Fig. 1 in the reaction mixture.
Fig. 2. Effect of the reaction time on the epoxidation of cyclohexene catalyzed by
MTO/C1. (1) MTO conversion, (2) MTO/C1 conversion, (3) MTO selectivity and (4)
MTO/C1 selectivity. Reaction conditions: cyclohexene 5 mmol, 30% H2O2 10 mmol,
MTO 0.05 mmol, C1 0.05 mmol, methanol 7 mL, reaction temperature 20 ◦C.
selectivity of epoxide decreased dramatically with time in the case
of neat MTO-catalyzed epoxidation, and to 27% when the reaction
finished; however, the selectivity of epoxide decreased very slowly
and as high as 93% of the selectivity of epoxide was received at the
stage of reaction completion in case of C1 as an additive. The results
displayed in Fig. 2 clearly illustrated that C1 suppressed the epoxide
ring-opening in the MTO-catalyzed epoxidation of cyclohexene.
3.3. The lifetime of MTO/C1 catalytic system
To further certify the stabilization effect of ILs on MTO or its
mono and bisperoxo derivatives in the epoxidation process, fresh
cyclohexene and 30% H2O2 were introduced into the reaction mix-
ture after the completion of each catalytic reaction. The results are
summarized in Table 4. It can be seen from Table 4, that the con-
version of cyclohexene remained nearly constant in the first 5 runs,
though the selectivity decreased in some degree. The results indi-
cated that MTO or its mono and bisperoxo derivatives were stable
throughout the epoxidation process and the activity of MTO was
maintained in all the time due to the presence of C1. The gradual
decreasing of the selectivity of the epoxide is due to the accu-
mulation of water from 30% H2O2. In addition, epoxide generated
during the process was not separated from the reaction mixture
after each run, which increased the concentration of the epoxide
and the contact time of the epoxide with water. As a result, more
epoxide ring-opening was observed. With more and more water
was accumulated with the catalytic running times, the concen-
tration of H2O2 was lower and lower, therefore, the epoxidation
rate decreased gradually too. Anyway, the introduction of the ILs
is in favor of stabilizing the MTO and its active derivatives in the
Table 4
The lifetime of MTO/C1 catalytic system.
3.2.4. Effect of the reaction time on the epoxidation of
cyclohexene catalyzed by MTO/C1
Entry
Running times
Time
Conversion (%)
Selectivity (%)
1
2
3
4
5
6
7
1
2
3
4
5
6
6
8
8
8
8
8
>99
>99
>99
>99
>99
85
>99
99
97
94
81
62
48
The effect of C1 on the epoxidation of cyclohexene at different
reaction time is shown in Fig. 2. It can be seen that the conver-
sion of cyclohexene with C1 as an additive increased with time
as expected, and reached about 100% in 3 h; while in case of neat
MTO catalyzed epoxidation of cyclohexene the reaction completed
within 1.5 h, indicating that the introduction of C1 decreased the
epoxidation rate. On the other hand, the introduction of C1 is favor-
able to increase the selectivity of epoxidation. One will see that the
8
11
>99
Reaction conditions: cyclohexene 5 mmol, 30% H2O2 10 mmol, MTO 0.05 mmol, C1
0.05 mmol, methanol 7 mL, reaction temperature 0 ◦C.

Y. Zhang et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 149–155
155
epoxidation process and the increasing of the selectivity of the
epoxidation.
[10] W.A. Herrmann, R.W. Fischer, M.U. Rauch, W. Scherer, J. Mol. Catal. 86 (1994)
243–266.
[11] M.M. Au-Omar, P.J. Hansen, J.H.J. Espenson, J. Am. Chem. Soc. 118 (1996)
4966–4974.
3.4. Recycling test of C1
[12] W.A. Herrmann, F.E. Kühn, Acc. Chem. Res. 30 (1997) 169–180.
[13] W.A. Herrmann, F.E. Kühn, M.U. Rauch, J.D.G. Correia, G. Artus, Inorg. Chem. 34
(1995) 2914.
By evaporating the reaction mixture at high vacuum, we tried to
recover both MTO and C1 from the MTO/C1 catalyzed epoxidation
of cyclohexene with aqueous H2O2 as an oxidant. When the residue
was subjected to epoxidation of cyclohexene directly, no reaction
was observed. However, when fresh MTO was introduced into the
reaction mixture, the epoxidation reaction proceeded smoothly
again. Analysis results showed that the conversion of cyclohexene
and the epoxide selectivity were same as those in the case of fresh
C1 as an additive. The results indicated that C1 was recyclable due
to its stability in the catalytic and recovery processes.
[14] W.A. Herrmann, F.E. Kühn, C.C. Romão, H. Tran Huy, M. Wang, R.W. Fischer, W.
Scherer, P. Kiprof, Chem. Ber. 26 (1993) 45.
[15] W.A. Herrmann, F.E. Kühn, M.R. Mattner, G.R.J. Artus, M.R. Geisberger, J.D.G.
Correia, J. Organomet. Chem. 538 (1997) 203–209.
[16] W.A. Herrmann, H. Ding, R.M. Kratzer, F.E. Kühn, J.J. Haider, R.W. Fischer, J.
Organomet. Chem. 549 (1997) 319–322.
[17] J. Rudolph, K.L. Reddy, J.P. Chiang, K.B. Sharpless, J. Am. Chem. Soc. 119 (1997)
6189–6190.
[18] C. Copéret, H. Adolfsson, K.B. Sharpless, Chem. Commun. 16 (1997) 1565–1566.
[19] W.A. Herrmann, J.D.G. Correia, M.U. Rauch, G.R.J. Artus, F.E. Kühn, J. Mol. Catal.
A: Chem. 118 (1997) 33–45.
[20] W.D. Wang, J.H. Espenson, J. Am. Chem. Soc. 120 (1998) 11335–11341.
[21] W.A. Herrmann, R.M. Kratzer, H. Ding, W.R. Thiel, H. Glas, J. Organomet. Chem.
555 (1998) 293–295.
[22] F.E. Kühn, A.M. Santos, P.W. Roesky, E. Herdtweck, W. Scherer, P. Gisdakis, I.V.
Yudanov, C. Di Valentin, N. Rösch, Chem. Eur. J. 5 (1999) 3603–3615.
[23] F.E. Kühn, A.M. Santos, I.S. Goncalves, C.C. Romão, A.D. Lopes, Appl. Organomet.
Chem. 15 (2001) 43–50.
[24] H. Adolfsson, A. Converso, K.B. Sharpless, Tetrahedron Lett. 40 (1999)
3991–3994.
[25] H. Adolfsson, C. Copéret, J.P. Chiang, A.K. Yudin, J. Org. Chem. 65 (2000)
8651–8658.
[26] P. Ferreira, W.D. Xue, É. Bencze, E. Herdtweck, F.E. Kühn, Inorg. Chem. 40 (2001)
5834–5841.
[27] M.J. Sabater, M.E. Domine, A. Corma, J. Catal. 210 (2002) 192–197.
[28] C.D. Nunes, M. Pillinger, A.A. Valente, I.S. Gonc¸ alves, J. Rocha, P. Ferreira, F.E.
Kühn, Eur. J. Inorg. Chem. (2002) 1100–1107.
[29] S.M. Nabavizadeh, Inorg. Chem. 42 (2003) 4204–4208.
[30] S.M. Nabavizadeh, M. Rashidi, J. Am. Chem. Soc. 128 (2006) 351–357.
[31] M.D. Zhou, J. Zhao, J. Li, S. Yue, C.N. Bao, J. Mink, S.L. Zang, F.E. Kühn, Chem. Eur.
J. 13 (2007) 158–166.
[32] S. Yamazaki, Tetrahedron 64 (2008) 9253–9257.
[33] M.D. Zhou, S.L. Zang, E. Herdtweck, F.E. Kühn, J. Organomet. Chem. 693 (2008)
2473–2477.
[34] A. Capapé, M.D. Zhou, S.L. Zang, F.E. Kühn, J. Organomet. Chem. 693 (2008)
3240–3244.
4. Conclusions
The four ionic liquids with pyridine N-oxide moiety as designed
and synthesized showed excellent performances in the MTO-
catalyzed epoxidation of olefins with 30% H2O2 as an oxidant. The
catalytic results displayed that low loading amount of the ILs com-
pared to other substances as additives could suppress the epoxide
ring-opening, which led to the significant increase of selectivity of
the epoxide. Though the introduction of the ILs caused the decrease
of the epoxidation rate, the side effect was much smaller than those
of other substances being used as additives. We concluded that the
coexistence of pyridine N-oxide and imidazolium moieties in the
ILs was important to deduce of the acidity of the central Re atom
meanwhile stabilize MTO or its derivatives in the epoxidation pro-
cess, which led to the excellent performances in improvement of
the selectivity of the epoxidation and the lifetime of MTO. It was
also displayed that the improvement degree on the selectivity of
epoxidation depended on the type of anion of the ILs, but not the
position of the substituent with imidazolium moiety in the ring of
pyridine N-oxide.
[35] Z.Q. Xu, M.D. Zhou, M. Drees, H. Chaffey-Millar, E. Herdtweck, W.A. Herrmann,
F.E. Kühn, Inorg. Chem. 48 (2009) 6812–6822.
[36] M.D. Zhou, Y. Yu, A. Capapé, K.R. Jain, E. Herdtweck, X.R. Li, J. Li, S.L. Zang, F.E.
Kühn, Chem. Asian J. 4 (2009) 411–418.
[37] P. Altmann, M. Cokoja, F.E. Kühn, J. Organomet. Chem. 701 (2012) 51–55.
[38] C.J. Qiu, Y.C. Zhang, Y. Gao, J.Q. Zhao, J. Organomet. Chem. 694 (2009)
3418–3424.
Acknowledgements
[39] Y. Gao, Y.C. Zhang, J.Q. Zhao, Chin. J. Catal. 30 (2009) 1243–1247.
[40] Y. Gao, Y.C. Zhang, C.J. Qui, J.Q. Zhao, Appl. Organomet. Chem. 25 (2011) 54–60.
[41] C.P. Ding, Y.C. Zhang, C.L. Wang, J.Q. Zhao, Chin. J. Mol. Catal. 24 (2010) 392–399.
[42] G.S. Owens, M.M. Abu-Omar, Chem. Commun. 116 (2000) 5–1166.
[43] D. Betz, W.A. Herrmann, F.E. Kuhn, J. Organomet. Chem. 694 (2009) 3320–3324.
[44] R. Saladino, R. Bernini, V. Neri, C. Crestini, Appl. Catal. A: Gen. 360 (2009)
171–176.
[45] R.Y. Yang, Y.C. Zhang, J.Q. Zhao, Catal. Commun. 12 (2011) 923–926.
[46] A. Mckillop, D. Kemp, Tetrahedron 45 (1989) 3299–3306.
[47] S. Carda-Broch, A. Berthod, D.W. Armstrong, Anal. Bioanal. Chem. 375 (2003)
191–199.
[48] F.E. Kühn, A. Scherbaum, W.A. Herrmann, J. Organomet. Chem. 689 (2004)
4149–4164.
[49] R.P. Swatloski, J.D. Holbrey, R.D. Rogers, Green Chem. 5 (2003) 361–363.
[50] L. Gubicza, N. Nemestóthy, T. Fráterb, K. Bélafi-Bakó, Green Chem. 5 (2003)
236–239.
This work was supported by the NSFC of China (grant no.
20776035, 20806020) and the NSF of Hebei Province of China (grant
no. B 2009000008).
References
[1] R.A. Sheldon, Green Chem. 9 (2007) 1273–1283.
[2] R. Noyori, M. Aoki, K. Sato, Chem. Commun. 16 (2003) 1977–1986.
[3] J.M. Brégeault, Dalton Trans. 17 (2003) 3289–3302.
[4] B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457–2473.
[5] S. Yamazaki, Org. Biomol. Chem. 8 (2010) 2377–2385.
[6] J. Piera, J.E. Bäckvall, Angew. Chem. Int. Ed. 47 (2008) 3306–3523.
[7] G. Grigoropoulou, J.H. Clark, J.A. Elings, Green Chem. 5 (2003) 1–7.
[8] W.A. Herrmann, M. Wang, Angew. Chem. Int. Ed. 103 (1991) 1709–1711.
[9] W.A. Herrmann, R.W. Fischer, D.W. Marz, Angew. Chem. Int. Ed. 30 (1991)
1638–1641.
[51] J.G. Shen, H. Wang, H.C. Liu, Y. Sun, Z.M. Liu, J. Mol. Catal. A: Chem. 28 (2008)
24–28.
[52] A. Chaumont, G. Wipff, J. Mol. Liq. 131-132 (2007) 36–47.