Mesostructured dihydroxy-functionalized guanidinium-based polyoxometalate with enhanced heterogeneous catalytic activity in epoxidation

Article abstract of DOI:10.1002/cplu.201300105

A mesostructured ionic liquid-polyoxometalate (IL-POM) hybrid has been prepared through designing a new dihydroxy-tethered guanidinium-based IL, N′′-(2,3-dihydroxypropyl)-N,N,N′,N′- tetramethylguanidinium chloride, to interact with Keggin-type POM phosphotungstic acid (H₃PW) in a self-assembly process. Scanning electron microscopy and transmission electron microscopy showed its special coral-shaped micromorphology. Nitrogen sorption analysis indicated the formation of a porous structure with a moderate surface area of about 30 m² g^-1 and narrowly distributed pore size located in the mesoscale. Assessed in the cis-cyclooctene epoxidation with H₂O₂, the mesostructured hybrid exhibited superior heterogeneous catalytic activity and steady reusability, and the conversion was more than four times that of homogeneous H₃PW itself, and more than 14 times that of the nonporous analogues. On the basis of the experimental results, a unique substrate-solvent-catalyst synergistic mechanism is proposed and discussed for understanding the dramatically enhanced catalytic performance. Copyright

Full text of DOI:10.1002/cplu.201300105

CHEMPLUSCHEM
FULL PAPERS
DOI: 10.1002/cplu.201300105
Mesostructured Dihydroxy-Functionalized Guanidinium-
Based Polyoxometalate with Enhanced Heterogeneous
Catalytic Activity in Epoxidation
[
a]
Guojian Chen, Yu Zhou, Pingping Zhao, Zhouyang Long, and Jun Wang*
A
mesostructured ionic liquid–polyoxometalate (IL-POM)
the mesoscale. Assessed in the cis-cyclooctene epoxidation
hybrid has been prepared through designing a new dihydroxy-
tethered guanidinium-based IL, N’’-(2,3-dihydroxypropyl)-
N,N,N’,N’-tetramethylguanidinium chloride, to interact with
with H O , the mesostructured hybrid exhibited superior heter-
2
2
ogeneous catalytic activity and steady reusability, and the con-
version was more than four times that of homogeneous H PW
3
Keggin-type POM phosphotungstic acid (H PW) in a self-assem-
itself, and more than 14 times that of the nonporous ana-
logues. On the basis of the experimental results, a unique
“substrate–solvent–catalyst” synergistic mechanism is pro-
posed and discussed for understanding the dramatically en-
hanced catalytic performance.
3
bly process. Scanning electron microscopy and transmission
electron microscopy showed its special coral-shaped micro-
morphology. Nitrogen sorption analysis indicated the forma-
tion of a porous structure with a moderate surface area of
2
À1
about 30 m g and narrowly distributed pore size located in
Introduction
Polyoxometalates (POMs), a family of structurally diverse transi-
tion-metal oxygen anion clusters, have achieved wide applica-
Ionic liquids (ILs) have received increasing attention as reac-
[6]
tion media/catalysts in organic syntheses, as well as tem-
plates or building blocks for constructing inorganic–organic
[1]
tions in catalysis, materials science, biology, and magnetism.
[7]
Intrinsically possessing controllable redox properties, various
POMs have been used extensively as efficient homogeneous
catalysts in many oxidative reactions, including the epoxida-
tion of olefins to epoxides, which are important intermediates
hybrid materials. Further, ILs have been used as adjustable
counter cations of POM cluster anions to obtain a family of IL-
POM hybrids applied in electrochemistry, catalysis, materials
[8]
science, and so forth. In the field of catalysis, several IL-POMs
have been reported as efficient homogeneous or phase-sepa-
[
2]
in fine-chemical and polymer synthesis. Nonetheless, pure
POMs are soluble in many polar solvents, which causes difficul-
ty in separation and recycling of the catalysts; therefore, many
efforts have been made to prepare heterogeneous POM-based
catalysts, such as immobilization of POMs on various porous
supports, for example, silica-based matrices and organic poly-
[9,10]
ration catalysts for epoxidation.
In addition, our group has
developed a “task-specific” IL-POM series through elaborate
[11]
functionalized IL cations, and most of them exhibit superior
catalytic performance in H O -based liquid-phase organic oxi-
2
2
dation reactions. Nonetheless, little research has been directed
towards the morphology and pore structure control of these
hybrid catalysts. The difficulty is how to design suitable ILs
that can self-assemble with the POM anion to create and con-
trol the porosity and accompanying morphology.
[
3]
mers, and assembly of POMs with appropriate organic coun-
[
4]
ter cations. The former favors higher surface areas to acceler-
ate the mass transfer, and the latter is beneficial for adjusting
the solubility, redox properties, and surface microenvironment
of POM catalysts through modifying POMs with diverse organic
cations. However, the self-assembled POM hybrids are fre-
Among numerous ILs, guanidinium-based ILs (GILs) are of
great interest for self-assembly synthesis, because the flexible
tunability of guanidinium provides great opportunities for pre-
paring a diverse range of cationic structures, and the planar
C=N p-conjugated structure exhibits excellent thermal and
[
5]
quently nonporous, thus limiting the heterogeneous catalytic
activity in mass-transfer-controlled systems. Therefore, it is nec-
essary to develop new methods to generate suitable micro/
mesostructures on these POM materials.
[12]
robust chemical stabilities. However, previous reports have
only introduced polyethylene glycol/ether, ÀSO H, and olefinic
3
units as the functional groups to modify guanidinium cat-
[
a] G. Chen, Dr. Y. Zhou, P. Zhao, Z. Long, Prof. Dr. J. Wang
State Key Laboratory of Materials-Oriented Chemical Engineering
College of Chemistry and Chemical Engineering
Nanjing University of Technology
Nanjing, Jiangsu 210009 (P. R. China)
Fax: (+86)25-83172264
[13]
ions, thus leaving room to develop new task-specific GILs for
creating IL-POM hybrid catalysts.
Herein, a new dihydroxy-functionalized guanidinium IL,
N’’-(2,3-dihydroxypropyl)-N,N,N’,N’-tetramethylguanidinium
chloride ([TMGDH]Cl), is designed to fabricate a mesoporous
IL-POM hybrid through ion exchange with the commercial
Keggin-type POM H PW O (H PW=phosphotungstic acid;
E-mail: junwang@njut.edu.cn
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201300105.
3
12 40
3
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 561

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Scheme 1. Synthesis of dihydroxy-tethered guanidinium-based polyoxo-
metalate [TMGDH]2.3
nidinium chloride.
H0.7PW. [GlTMG]Cl=N’’-glycidyl-N,N,N’,N’-tetramethylgua-
Figure 1. A,B) SEM and C,D) HRTEM images of [TMGDH]2.3H0.7PW.
Scheme 1). The dihydroxy group is introduced for two purpos-
es. One is to control the morphology and pore structure of the
obtained IL-POM hybrid. The other is to improve the catalytic
performance, because it has been reported for epoxidation of
olefins with H O that introducing hydroxyl groups onto the
(0.4 nm) to contribute any additional surface area when tested
by nitrogen physisorption.
2
2
surface of heterogeneous catalysts can enhance the catalytic
During the synthesis of the catalysts, it was found that the
dihydroxy-tethered group in the IL cation plays a key role in
the formation of the above special morphology. For example,
similar primary particles and micromorphology along with
[
14]
activity, and the hydrogen bonding involved in the active
metal-peroxo species in the presence of protic solvents has
[
15]
been regarded as a crucial catalytic promoter. The catalyst is
1
13
characterized by H NMR, C NMR, FTIR, UV/Vis, and ESR spec-
troscopy, SEM, TEM, XRD, thermogravimetry (TG), nitrogen
sorption, and elemental analysis,
clear channels and cavities are observed in [TMGDH] PW (Fig-
3
ure 2A), which is also prepared by employing the same IL
and its heterogeneous catalytic
performance is assessed in the
liquid-phase epoxidation of cis-
cyclooctene with H O . Various
2
2
control catalysts are also pre-
pared and tested for compari-
son.
A unique “substrate–sol-
vent–catalyst” mechanism is pro-
posed to understand the superi-
or catalytic performance of the Figure 2. SEM images of A) [TMGDH] PW, B) [TMGOH]2.2H0.8PW, and C) [TMG] PW.
3
3
material.
[
TMGDH]Cl but with Na PW; nonetheless, the packing of pri-
3
Results and Discussion
mary particles is closer than for [TMGDH] H PW. On the con-
2.3 0.7
trary, the morphology changes to smooth angular blocks with
a size in micrometers by using monohydroxy-tethered and hy-
droxy-free GIL cations to prepare [TMGOH] H PW and
Formation of mesostructured GIL-POM catalysts
The mesoporous dihydroxy-tethered GIL-POM catalyst, denot-
ed as [TMGDH] H PW, is prepared through the reaction be-
2.2 0.8
[TMG] PW hybrids (Figure 2B,C), and no substructure can be
2.3 0.7
3
tween [TMGDH]Cl and H PW in aqueous solution.
identified.
3
SEM images (Figure 1A,B) indicate that the primary particles
of [TMGDH] H PW have a flexible noodle shape with a diame-
The surface area and pore structure are further characterized
by nitrogen sorption experiments. As shown in Figure 3A,
curve a, the nitrogen sorption isotherm of [TMGDH] H PW is
2.3 0.7
ter of about 20 nm and a length of tens of nanometers, and
these noodles are intertwined with each other to form a netlike
substructure (Figure 1A) with a nanoscale hollow structure and
irregular coral-shaped morphology at the micrometer level.
Figure 1C and D show high-resolution TEM (HRTEM) images of
2.3 0.7
type IV with a clear H1-type hysteresis loop in the partial pres-
sure range of P/P =0.5–0.9, characteristic of a mesoporous
0
material. Curve a in Figure 3B gives the pore size distribution
of [TMGDH] H PW calculated by the Barrett–Joyner–Halenda
2.3 0.7
[
TMGDH] H PW, which confirm the existence of random mes-
method, which shows a narrow pore size distribution with the
most probable pore size of 6.2 nm, in agreement with the
result of the TEM image. Table 1 lists the surface areas, pore
2
.3 0.7
opores in coral-shaped particles and micropores in the primary
particles. However, the size of these micropores is too small
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 562

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Figure 4. XRD patterns of A) a) H
B) c) [TMGDH] PW, d) [TMGOH]2.2
3
PW12
O
40, b) [TMGDH]2.3
H0.7PW;
Figure 3. A) N
a) [TMGDH]2.3
2
sorption isotherms and B) pore size distribution of
0.7PW and b) [TMGDH] PW.
3
H
0.8PW, and e) [TMG]
3
PW.
H
3
18.5 nm (Figure 3B, curve b).
Table 1. Textural properties and catalytic performances of POM catalysts in the epoxidation of cis-cyclooctene
However, the other samples pre-
pared from monohydroxy-teth-
ered and hydroxy-free GIL cat-
ions all exhibit very small surface
area, even lower than that of
[
d]
2 2
with H O .
[
a]
[b]
[c]
[e]
[f]
Entry
Catalyst
S
[
BET
V
p
D
av
Solubility
in reaction
Conversion
[%]
Selectivity
[%]
2
À1
3
À1
H PW itself (Table 1).
3
m g
]
[cm g
]
[nm]
Figure 4 shows the XRD pat-
terns of various GIL-POM hy-
brids. The XRD pattern of
H PW O presents a number of
1
2
3
4
5
6
7
[TMGDH]Cl
–
–
–
0.06
0.14
–
–
–
10.4
25.5
–
soluble
soluble
0
–
84.7
98.5
99.0
100
100
31.6
H
3
PW
[TMGDH]2.3
[TMGDH] PW
[TMGOH]2.2
[TMG] PW
30%H PW/SiO
5.4
28.0
22.6
2.1
0.3
204
22.3
99.5
96.5
6.5
7.0
74.4
H
0.7PW
insoluble
insoluble
insoluble
insoluble
insoluble
3
3
12 40
H
0.8PW
sharp Bragg peaks, characteristic
of Keggin-type phosphotung-
3
–
0.56
–
7.0
[
g]
3
2
[16]
state (PW),
but these peaks
[
a] BET surface area. [b] Total pore volume. [c] Average pore size. [d] Reaction conditions: cis-cyclooctene
10 mmol), 30 wt% H (, 5 mmol), catalyst (0.1 g), solvent (methanol, 3 mL), 608C, 6 h. [e] Conversion based
. [f] Selectivity for 1,2-epoxycyclooctane; by-products: 2-cycloocten-1-ol, 2-cycloocten-1-one, and 1,2-cy-
clooctanediol. [g] The amount of 30% H PW/SiO was 0.3 g.
disappear after combining with
(
2
O
2
+
IL cation [TMGDH] (Figure 4A).
2 2
on H O
For
TMGDH] PW samples,
[TMGDH] H PW
and
3
2
2
.3 0.7
[
a
new
3
broad Bragg reflection is visible
at 2q=8.28, with a d spacing of 1.08 nm, close to the size of
volumes, and average pore sizes. The results indicate that
[17]
[
TMGDH] H PW has
a
moderate BET surface area of
a basic structural unit of IL-POM. This phenomenon is in ac-
cordance with the result of the HRTEM image (Figure 1D), in
which the primary structural units (ca. 1.1 nm) are distributed
homogeneously throughout the IL-POM secondary structures
to form uniform wormlike small micropores, which correspond
2
À1
.3 0.7
2
2
8 m g , close to that of the previously reported polyammo-
nium-POM hybrid mesoporous materials or a peroxophospho-
tungstate-based polymer-immobilized IL-phase catalyst with
2
À1 [3c,4a]
surface area of 27–51 m g .
Also, the sample
[18]
[
TMGDH] PW prepared from the same IL as [TMGDH] H PW
to the residual voids of ion-pair assembly. In addition, many
new characteristic peaks appear in the XRD patterns of
[TMGOH] H PW and [TMG] PW (Figure 4B), thus indicating
3
2.3 0.7
shows a type IV nitrogen sorption isotherm with H1-type hys-
teresis loop in a relatively higher partial pressure range of P/
2.2 0.8
3
P =0.8–0.99, which corresponds to
a
surface area of
the formation of new three-dimensional crystal structures as-
0
2
À1
[11d,19]
22.6 m g (Table 1) and a larger pore size most probably of
sembled by Keggin anions with GIL cations,
in good
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 563

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
agreement with the polyhedral morphologies of their SEM
images.
dition, several new peaks emerge on the spectrum of
À1
[TMGDH] H PW. The weak band at 1053 cm is a branched
2.3 0.7
À1
peak from the stretching PÀO vibration band at 1079 cm ,
which implies the distortion of the symmetric Keggin frame-
work owing to the partial degradation of the pH-sensitive
Further characterization of [TMGDH] H PW
2.3 0.7
3
À
PW O
40
[22]
anion in IL aqueous solution.
Another weaker
12
The hybrid [TMGDH] H PW was further characterized by ele-
2
.3 0.7
À1
1
13
band at 951 cm is assigned to the split peak from the vibra-
mental analysis, TG, and H NMR, C NMR, FTIR, UV/Vis, and
ESR spectroscopy. The elemental analysis found (wt%) C 6.79,
N 2.90, and H 1.56, close to the calculated values of C 6.67,
N 2.92, and H 1.39, thus indicating the formula of
VI
À1
tion of W =O at 979 cm , which suggests the existence of re-
5
+
duced W species as a result of the intramolecular charge
VI
transfer from the p-conjugated TMG moiety to terminal W =
[23]
O.
Other new peaks at 2964, 1618, 1560, 1472, and
[
TMGDH] H PW shown in Scheme 1. The structure of the
2.3 0.7
À1
[24]
1
13
1410 cm are assigned to the TMG moiety. The strong and
sample was further confirmed by H and C NMR spectroscopy
see Figure S1 in the Supporting Information). Particularly, in
À1
broad peak at 3432 cm is assigned to the stretching vibration
(
+
1
of OÀH in [TMGDH] , thereby indicating the existence of mul-
the H NMR spectra, the signals for [TMGDH]Cl at d=6.66 (OH)
and 8.11 ppm (NH) shift to d=6.23 (OH) and 7.78 ppm (NH)
for [TMGDH] H PW, respectively, thus implying the existence
[25]
tiple hydrogen-bonding networks in the IL-POM hybrid. The
hydrogen-bonding interactions may account for the solid
nature of [TMGDH] H PW with a high melting point of ap-
2.3 0.7
2.3 0.7
of active protons in the latter sample. The above phenomenon
strongly demonstrates that the counter protons of the POM
anion have not been totally exchanged by GIL cations in the
ion-exchange process, probably because of the strong acidic
proximately 2508C. In addition, [TMGDH] H PW is insoluble
2.3 0.7
in most common solvents, including water, alcohols, ethyl ace-
tate, acetic acid, acetone, acetonitrile, and DMF, and is only
sparingly soluble in DMSO.
[
11c]
environment relative to H PW O .
The TG curve (see Fig-
3
12 40
Figure 6 depicts the UV/Vis and ESR spectra of
ure S2 in the Supporting Information) indicates a thermally
[
TMGDH] H PW and pure H PW O , which is applied to re-
2.3 0.7 3 12 40
stable structure for [TMGDH] H PW up to 2408C. The slight
2.3 0.7
flect the electronic behavior of catalysts. In the UV/Vis spec-
trum of [TMGDH] H PW, a broad absorption band in the
weight loss at the early heating stage below 2008C results
from the release of moisture and constitutional water, whereas
the drastic weight loss in the range 240–6008C is assigned to
the decomposition of the hydroxy-attached TMG moiety plus
the complete collapse of the inorganic Keggin PW structure
2.3 0.7
range of 600–800 nm is assignable to the intramolecular
6
+
5+
[26]
charge transfer from W to W (Figure 6A, curve a), but
the band was undetectable for H PW O (Figure 6A, curve b).
3
12 40
[
20]
The ESR signals at 3500 G on curve a in Figure 6B also reveal
with the formation of P O and WO . Moreover, the total
2
5
3
5
+
the existence of the low-valent W species in the hybrid
weight loss of about 13.30% in the range 240–6008C is very
close to the value of 13.20% calculated theoretically from ele-
mental analysis, once again verifying the chemical composition
of [TMGDH] H PW.
2.3 0.7
Figure 5 shows the FTIR spectra of H PW O
and
3
12 40
[
TMGDH] H PW. The spectrum of H PW O possesses vibra-
2
.3 0.7
3
12 40
tion bands characteristic of the Keggin anion located at 1080,
À1
9
83, 889, and 803 cm , corresponding to vibrations n(PÀO ),
a
[
21]
n(W=O ), n(WÀO ÀW), and n(WÀO ÀW), respectively.
None-
t
b
c
theless, these four characteristic peaks for the Keggin structure
occur with light shifts on [TMGDH] H PW, which indicates
2.3 0.7
+
strong ionic interactions between the [TMGDH]
and
3
À
PW O
40
and thus the formation of the IL-POM hybrid. In ad-
12
Figure 6. A) UV/Vis and B) ESR spectra of a) [TMGDH]2.3
b) H
H0.7PW and
Figure 5. FTIR spectra of a) H
3
PW12
O
40 and b) [TMGDH]2.3
H0.7PW.
3 40
PW12O .
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 564

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
5
+
[
TMGDH] H PW, and curve b indicates the absence of W
2.3 0.7
species in pure H PW O . These phenomena plus the results
3
12 40
5
+
from IR spectra sufficiently confirm the formation of W spe-
cies during the ion exchange of [TMGDH]Cl and H PW O .
3
12 40
Catalytic activity in the epoxidation reaction
The catalytic performances of various catalysts are assessed in
the epoxidation of the probe substrate cis-cyclooctene with
aqueous H O as oxidant and methanol as solvent. As shown
2
2
in Table 1, the neat precursors IL [TMGDH]Cl and POM H PW
3
dissolve in the reaction medium to give a homogeneous
system, but their catalytic activities are very low (Table 1, en-
tries 1 and 2): no product is detected by using [TMGDH]Cl, and
low conversion of 22.3% and selectivity of 84.7% are observed
by using H PW. In contrast, the hybrid catalyst
3
[
TMGDH] H PW is insoluble and well dispersed in the oxida-
2.3 0.7
tive reaction medium, thus leading to a heterogeneous system
and exhibiting a high conversion of 99.5% and selectivity of
98.5% (Table 1, entry 3) under the optimized reaction condi-
tions. The heterogeneous catalyst [TMGDH] H PW can be re-
2.3 0.7
covered simply by filtration after reaction, and the catalytic re-
usability was measured by a five-run test without adding any
fresh catalyst after each run. As shown in Figure S3, there was
no significant loss of conversion and selectivity after five runs,
which implied good reusability of this catalyst.
Figure 7. A) Catalytic performance in epoxidation of cis-cyclooctene with
H
2
O
2
in different solvents. B) Influence of different molar ratios of methanol
On the contrary, nonporous catalysts [TMGOH] H PW and
2.2 0.8
to cis-cyclooctene (nmet/ncyc). Reaction conditions: cis-cyclooctene (10 mmol),
[
TMG] PW with very small surface areas all give poor conver-
3
30 wt% H
2 2
O (5 mmol), catalyst [TMGDH]2.3H0.7PW (0.1 g), solvent (3 mL),
sions of 6.5 and 7.0%, respectively (Table 1, entries 5 and 6),
608C, 6 h. Con.=conversion, Sel.=selectivity.
even lower than that of homogeneous H PW. The above phe-
3
nomenon suggests that the morphology and porous structure
play an important role in this reaction. To support this deduc-
[
28]
POM-catalyzed epoxidation reactions of olefins. Nonetheless,
modest conversion of 84.6% and low conversion of 28.5% are
obtained by acetonitrile and ethyl acetate, respectively. Inter-
estingly, high conversions are obtained by using short-chain
aliphatic alcohol solvents. Among them, methanol provides
the highest conversion of 99.5% and selectivity of 98.5%. On
increasing the carbon chain length of aliphatic alcohols, the
conversions continuously decrease to 93.4% for ethanol,
89.5% for 1-propanol, and 88.6% for tert-butyl alcohol. Note
that these short-chain alcohols are hardly oxidized by using
tion, the proton-free counterpart [TMGDH] PW with similar
3
morphology and mesostructure to [TMGDH] H PW was
2.3 0.7
tested, which also offered comparably a high conversion of
6.5% and selectivity of 99.0%. These results indicate that the
9
coral-shaped morphology and accompanying mesostructure
favor a better catalytic performance. Moreover, this control ex-
periment also reveals that the protonation of [TMGDH] H PW
2.3 0.7
is not the key contribution to the superior catalytic perfor-
[
11c]
mance, little different from that in the previous report.
Nota-
[
29]
bly, the self-formed mesostructure in [TMGDH] H PW or
PW anions as catalysts in mild conditions, thus avoiding the
2.3 0.7
[30]
[
TMGDH] PW possesses unique advantages for the epoxidation
possible role of “sacrificial” solvents in our reaction. In addi-
tion, a common polar protic solvent, acetic acid, was em-
ployed, which achieved a good conversion of 98.0% and selec-
tivity of 99.0%, whereas a typical aprotic solvent, acetone, only
gives a very low conversion of 15.6%. According to the above
3
reaction. For example, 30% H PW/SiO obtained through di-
3
2
rectly impregnating H PW on the hydroxy-enriched porous
3
silica support only gives a moderate conversion of 74.4% but
with a low selectivity of 31.6% (Table 1, entry 7), owing to the
hydration of epoxide to diol in the protonated hydroxy acidic
[31]
results and referring to Kamlet–Taft plots, it is found that
“greener” protic solvents with moderate polarity facilitate
good performance in our reaction system.
[27]
conditions and allylic oxidation in the homolytic pathway.
The influence of the solvent amount was also tested with
methanol. As shown in Figure 7B, the conversion is only 23.4%
with low selectivity of 68.2% in the solvent-free reaction.
When methanol is used as solvent, both conversion and selec-
tivity gradually increase along with the enhancement of the
Solvent effects in the epoxidation reaction
As shown in Figure 7A, the catalyst [TMGDH] H PW exhibits
2.3 0.7
different performances in the epoxidation of cis-cyclooctene
with H O by using various solvents. During the reaction, the
2
2
catalyst is insoluble in all the solvents. Acetonitrile and ethyl
acetate are the two commonly used good solvents in many
molar ratio of methanol to cis-cyclooctene (nmet/n ), thereby
cyc
achieving the highest value as nmet/ncyc stays on 8. Under these
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 565

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
3
À
conditions, the reaction is
a
liquid–liquid–solid triphasic
W sites to create the {PO [WO(O ) ] } -containing tungsten-
4 2 2 4
[33b]
system, because such a low amount of methanol is unable to
completely homogenize the organic phase of cis-cyclooctene.
On further increasing the methanol amount, the reaction
system tends to a liquid–solid biphasic state owing to the
gradual solution of cis-cyclooctene in methanol; however, the
conversion decreases though the selectivity is unchanged.
When the value of nmet/ncyc is 14, the amount of methanol is
enough to dissolve all the cis-cyclooctene to give a liquid–solid
heterogeneous system; but the conversion is only 92.4%. In
previous works, a large volume of solvent was usually used for
peroxo complex II.
Afterwards, the substrate cis-cyclooctene
attacks the tungsten-peroxo species in II to generate the possi-
ble tungsten-peroxo–cyclooctene intermediate structure III,
then gives the corresponding epoxide product along with the
final regeneration of the original catalyst. Different from a ho-
mogeneous reaction, mass transfer is an important factor in
the heterocatalytic process, involving the influences of mor-
phology and pore structure, and the substrate–solvent–catalyst
interaction.
As described in Scheme 2, several steps are related to the
mass-transfer process, for example, the attack of complex I to
W sites and the accessing process of cis-cyclooctene to the
tungsten-peroxo species in II. In these processes, the behavior
of the catalytically active sites is limited by the mass-transfer
efficiency, and thus the morphology and mesostructure play
a key role in the reaction. Generally, loosely packed nanoparti-
cles with moderate mesopores favor fast mass transfer and the
exposure of catalytically active sites on internal surfaces. There-
[
10a,32]
homogenizing substrates;
yet, in our experiment, a special
phenomenon is observed that a relatively small solvent/sub-
strate molar ratio favors higher conversion.
Substrate–solvent–catalyst mechanism
Based on the above results and previous studies on epoxida-
tion of olefins, a “substrate–solvent–catalyst” reaction mecha-
nism is proposed in Scheme 2 to understand the superior per-
fore, the mesostructured samples ([TMGDH]2.3
H0.7PW and
[
TMGDH] PW) with coral-shaped
3
morphology all exhibit superior
catalytic performances (Table 1).
In detail, cis-cyclooctene and
H O molecules in methanol sol-
2
2
vent diffuse easily into the inter-
nal surface of the solid catalyst
through the mesopores, and the
nanocavities among POM anions
allow good access of substrates
and H O2 to W species in the
2
POM anion, which accelerates
mass transfer and the formation
of the catalytically active species
tungsten-peroxo complex II. On
the contrary, the catalytic epoxi-
dation activities of nonporous
samples ([TMGOH] H PW and
Scheme 2. Proposed catalytic mechanism for [TMGDH]2.3
in solvent CH OH.
2 2
H0.7PW-catalyzed epoxidation of cis-cyclooctene with H O
2.2 0.8
3
[TMG] PW) with bulky block-
3
shaped morphology are always
very low (Table 1), in spite of
formance of the catalyst [TMGDH] H PW. Previous works on
having crystal structures. Therefore, fabricating suitable micro/
mesostructures for IL-POM catalysts is crucial for superior cata-
lytic activity.
2.3 0.7
PW-based catalysts indicate that tungsten-peroxo complex is
the catalytically active site for epoxidation of olefins, and the
epoxidation process usually proceeded in three steps: 1) the
formation of tungsten-peroxo complex through the interaction
between PW species and H O ; 2) attack of the olefin substrate
The substrate–solvent–catalyst interaction is another impor-
tant factor that influences the mass transfer. In a hetero
system, the substrates and catalyst are all surrounded by the
solvent, and a suitable solvent benefits the access of the sub-
strate to the catalyst and the reduction of the activation barrier
through the substrate–solvent–catalyst interaction. Thus, the
conversion varies with different solvents. When protic polar
solvents such as methanol are used, strong hydrogen-bonding
interactions will form between solvent molecules and H O , as
2
2
by the tungsten-peroxo complex, to generate the epoxide
[
33]
product; and 3) regeneration of the PW species. Besides, the
solvent will take part in the formation of the peroxo complex
through involving the interaction between the substrate and
[
34]
H O . As shown in Scheme 2, the epoxidation of cis-cyclooc-
2
2
tene catalyzed by [TMGDH] H PW also undergoes three
2.3 0.7
2
2
steps. First, the five-membered ring structure I is formed
well as the introduced hydroxyl groups in the catalyst, which
would thus decrease the adsorption resistance and benefit the
access of the substrate to the active site in the catalyst. Be-
sides, many previous reports indicate that the hydrogen-bond-
through the hydrogen-bonding interaction between H O and
2
2
[
35]
the hydroxyl group of methanol. Then the H O molecules
2
2
incorporated in complex I contact the Keggin-framework
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 566

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
ing networks in the “CH OH–H O ” complex can accelerate the
ic activity of hydroxy-rich mesostructured [TMGDH] H PW
3
2
2
2.3 0.7
activation of H O in metal-based oxidations through reducing
hybrid in the epoxidation of cis-cyclooctene with H O in
2
2
2
2
[
35,36]
the catalytic activation barrier.
Therefore, H O has been
protic solvents.
2
2
preliminarily activated by the formation of hydrogen bonds in
structure I before attacking the W sites, and thus the solvent–
oxidant interaction may be responsible for the extremely high Conclusion
catalytic performance in the epoxidation of cis-cyclooctene
A novel mesostructured IL-POM hybrid has been self-assem-
bled by ion exchange of a dihydroxy-containing guanidinium
IL with phosphotungstic acid. The obtained hybrid material ex-
hibits superior heterogeneous catalytic activity and steady re-
usability in the epoxidation of cis-cyclooctene with H O , as
with H O when methanol (or other protic polar solvent such
2
2
as acetic acid) is used as solvent (Figure 7A). Furthermore,
[
14]
recent reports have revealed the promotional functionality
of the hydroxyl groups or functionalized protic molecules in-
troduced on the surface of heterogeneous catalysts, such as Ti-
COE-4 and Ti-SBA-15, for epoxidations with H O . For the
2
2
a result of the promotion of unusual morphology and pore
structure, together with a hydrogen-bonding-enriched micro-
environment surrounding the POM anion. Particularly, the con-
trollable introduction of hydroxyl groups into IL cations not
only favors the pore structure and morphology control of the
IL-POM hybrid but also benefits the accessibility of active sites
in POM-based catalysts.
2
2
above reasons, the abundant hydroxyl groups in
TMGDH] H PW are not only able to promote the accessibili-
[
2.3 0.7
ty through the solvent–catalyst interaction, but are also benefi-
cial for the formation and stability of tungsten-peroxo species
in catalytically active complex II.
Moreover, the solvent amount determines the concentration
of substrate, and the mass transfer of substrate seems not to
be hindered by a smaller amount of solvent and highly con-
centrated substrate. As shown in Figure 7B, a suitable amount
of methanol solvent (8.0 molar equiv with respect to cis-cyclo-
octene) is required to obtain the high activity of 99.5% in a tri-
phasic system. However, the catalytic activity is low in the ab-
sence of methanol, which is ascribed to the access difficulty of
cis-cyclooctene to the catalyst. A very small amount of metha-
nol (nmet/ncyc =2) readily leads to a dramatically increased con-
version of 87.6% owing to the enhancement of mass transfer.
Excess methanol will reduce the concentration of substrate,
and thus slow down the reactivity of cis-cyclooctene. The
above phenomenon can be explained by mass-transfer effi-
ciency and concentration effects by tuning a moderate
amount of methanol.
Experimental Section
Materials and methods
All the chemicals were of analytical grade and used as received.
FTIR spectra were recorded on a Nicolet 360 FTIR instrument (KBr
disks) in the region 4000–400 cm . Solid UV/Vis spectra were mea-
sured with a PE Lambda 950 spectrometer and BaSO was used as
an internal standard. ESR spectra were recorded on a Bruker EMX-
À1
4
1
13
10/12 spectrometer at the X-band. H and C NMR spectra were
measured with a Bruker DPX 500 spectrometer at ambient temper-
ature in [D ]DMSO by using TMS as internal reference. The CHN el-
6
emental analysis was performed on an elemental analyzer Vario EL
cube. Melting points were determined using a ꢁ4 digital micro-
scopic melting point apparatus with an upper limit of 2508C. TG
analysis was performed with an STA409 instrument in dry air at
Finally, it should be pointed out that intramolecular electron
À1
a heating rate of 108Cmin . XRD patterns were collected on a Bru-
VI
transfer from the GIL organic moiety to terminal W =O in POM
ker D8 Advance powder diffractometer using a Ni-filtered CuKa radi-
clusters is a prerequisite for high catalytic performance. This
special electronic behavior leads to the coexistence of terminal
ation source at 40 kV and 20 mA, from 5 to 808 with a scan rate of
À1
0
.28 s . SEM images were obtained on a Hitachi S-4800 field-emis-
6
+
5+
[23,37]
sion scanning electron microscope. The TEM images were obtained
by using a JEOL JEM-2010 (200 kV) TEM instrument. BET surface
areas were measured at the temperature of liquid nitrogen (77 K)
by using a Micromeritics ASAP2010 analyzer; the samples were de-
W
/W species
in [TMGDH] H PW, which should large-
2.3 0.7
ly favor the redox property of W species for catalyzing the ep-
oxidation reaction with H O based on the active tungsten-
2
2
peroxo complex II. Accordingly, the lack of favorable electron
transfer from organic cations to POM units can be an interpre-
tation for the low catalytic activity of cis-cyclooctene epoxida-
tion with H O over H PW/SiO and H PW, though H PW/SiO
À3
gassed at 1508C for 3 h to a vacuum of 10 Torr before analysis.
2
2
3
2
3
3
2
Catalyst preparation
has a large surface area and the mass-transfer resistance is
The mesoporous dihydroxy-functionalized GIL-POM catalyst, denot-
ed as [TMGDH] H PW, was prepared in three steps (Scheme 1).
weaker if using a homogeneous H PW catalyst. However, for
3
2.3 0.7
the nonporous IL-POM catalysts, such as [TMGOH] H PW and
2.2 0.8
[
TMG] PW, no matter whether they possess the reduced-state
First,
N’’-glycidyl-N,N,N’,N’-tetramethylguanidinium
chloride
3
5
+
W
species (see ESR spectra in Figure S4), they all give low
[GlTMG]Cl was synthesized by direct N-alkylation of N,N,N’,N’-tetra-
methylguanidine (TMG) with epichlorohydrin without solvent,
based on the previously described method with slight modifica-
catalytic activities, because the morphology and porous struc-
ture of a catalyst affect significantly the accessibility of the
H O to W sites to form catalytically active W-peroxo species.
[
38]
tion.
TMG (5.76 g, 50 mmol) and epichlorohydrin (5.55 g,
2
2
6
0 mmol) were added to a flask and stirred magnetically at 408C
Briefly, it is the loosely packed nanoparticles with moderate
for 24 h under a nitrogen atmosphere. The white precipitate was
isolated by filtration, washed with acetone (5ꢁ10 mL), and dried
under vacuum to obtain a hygroscopic light yellow solid [GlTMG]Cl
with a yield of 70%.
mesopores that provide a suitable situation, in which the hy-
drogen-bonding-assisted substrate–solvent–catalyst synergistic
catalytic mechanism can be applied to explain the high catalyt-
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 567

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Second, the dihydroxy-tethered IL N’’-(2,3-dihydroxypropyl)-
N,N,N’,N’-tetramethylguanidinium chloride [TMGDH]Cl was pre-
100%; selectivity of epoxide=(mmol epoxide)/(mmol cis-cyclooc-
tene converted)ꢁ100%. During reusability assessment, the catalyst
was separated by filtration, washed with ethanol three times, dried
in a vacuum at 808C for 6 h, and then used for the next run.
[39]
pared by ring opening of [GlTMG]Cl in hot water.
[GlTMG]Cl
(
1.0 g) was dissolved in water (20 mL), and the obtained mixture
was heated at reflux with stirring at 708C for 24 h. After the reac-
tion, water was removed by rotary evaporation and the product
was dried under vacuum for 12 h. [TMGDH]Cl was obtained as Acknowledgements
a quite viscous light yellow oil at room temperature. [TMGDH]Cl:
1
H NMR (300 MHz, [D ]DMSO, TMS): d=2.93 (s, 12H; N-CH ), 3.41–
6
3
We are grateful to the Key Program of the National Natural Sci-
ence Foundation of China (No. 21136005).
3
.75 (m, 5H; CH , CH), 5.01 (s broad, 1H; OH), 6.66 (s broad, 1H;
2
13
OH), 8.11 ppm (s, 1H; NH); C NMR (75.5 MHz, [D ]DMSO, TMS):
6
d=39.50 (N-CH ), 60.68 (CH ), 65.89–67.19 (CH ), 75.92 (CH),
3
2
2
1
60.90 ppm (C-N).
Keywords: epoxidation
· heterogeneous catalysis · ionic
liquids · morphology · polyoxometalates
Finally, [TMGDH] H PW was prepared by reacting [TMGDH]Cl
2
.3 0.7
(
3 mmol) with H PW O (1 mmol) in aqueous solution stirred at
3 12 40
[
[
1] D. Long, E. Burkholder, L. Cronin, Chem. Soc. Rev. 2007, 36, 105–121.
2] a) C. L. Hill, C. M. Prosser-McCartha, Coord. Chem. Rev. 1995, 143, 407–
room temperature for 24 h. The resulting light yellow precipitate
was isolated by filtration, washed with water, and dried in
a vacuum at 808C to give the final product with a yield of 94%.
455; b) N. Mizuno, K. Yamaguchi, K. Kamata, Coord. Chem. Rev. 2005,
2
49, 1944–1956.
1
[
TMGDH] H PW: H NMR (300 MHz, [D ]DMSO, TMS) (Figure S1A):
2
.3 0.7
6
[
3] a) K. Yamaguchi, C. Yoshida, S. Uchida, N. Mizuno, J. Am. Chem. Soc.
005, 127, 530–531; b) S. P. Maradur, C. Jo, D. H. Choi, K. Kim, R. Ryoo,
d=2.90 (s, 12H; N-CH ), 3.50–3.80 (m, 5H; CH , CH), 4.93 (s broad,
1
(
(
3
5
3
2
2
13
H; OH), 6.23 (s broad, 1H; OH), 7.78 ppm (s, 1H; NH); C NMR
ChemCatChem 2011, 3, 1435–1438; c) S. Doherty, J. G. Knight, J. R. Elli-
son, D. Weekes, R. W. Harrington, C. Hardacre, H. Manyar, Green Chem.
2012, 14, 925–929; d) M. Carraro, G. Fiorani, L. Mognon, F. Caneva, M.
Gardan, C. Maccato, M. Bonchio, Chem. Eur. J. 2012, 18, 13195–13202.
75.5 MHz, [D ]DMSO, TMS) (Figure S1B): d=39.50 (N-CH ), 52.09
6
3
CH ), 61.07 (CH ), 67.80 (CH), 160.91 ppm (C=N); FTIR (KBr): n˜ =
2 2
437, 2965, 1617, 1560, 1473, 1410, 1260, 1080, 979, 896, 808, 595,
À1
[4] a) M. V. Vasylyev, R. Neumann, J. Am. Chem. Soc. 2004, 126, 884–890;
b) A. Nisar, X. Wang, Dalton Trans. 2012, 41, 9832–9845.
[
20 cm . In addition, the unprotonated [TMGDH] PW was pre-
3
pared by similar procedures by anion exchange of [TMGDH]Cl with
Na PW O but with a low yield of 40%.
5] a) N. Mizuno, S. Uchida, K. Kamata, R. Ishimoto, S. Nojima, K. Yonehara,
Y. Sumida, Angew. Chem. 2010, 122, 10168–10172; Angew. Chem. Int.
Ed. 2010, 49, 9972–9976; b) S. Uchida, K. Kamata, Y. Ogasawara, M.
Fujita, N. Mizuno, Dalton Trans. 2012, 41, 9979–9983.
3
12 40
N’’-Ethanol-N,N,N’,N’-tetramethylguanidinium bromide [TMGOH]Br
was synthesized according to the previous literature.
(
solved in ethanol (20 mL) and heated at reflux with stirring at 808C
for 24 h. A colorless liquid was obtained by distilling the solvent
and it was then solidified at room temperature. The white solid
was washed thoroughly with diethyl ether (5ꢁ10 mL) and dried at
[40]
TMG
[
[
6] a) Q. Zhang, S. Zhang, Y. Deng, Green Chem. 2011, 13, 2619–2637;
b) J. P. Hallett, T. Welton, Chem. Rev. 2011, 111, 3508–3576.
7] a) Z. Ma, J. Yu, S. Dai, Adv. Mater. 2010, 22, 261–285; b) E. Ahmed, M.
Ruck, Angew. Chem. 2012, 124, 314–316; Angew. Chem. Int. Ed. 2012,
2.30 g, 20 mmol) and 2-bromoethanol (3.75 g, 30 mmol) were dis-
5
1, 308–309; c) H. Fu, C. Qin, Y. Lu, Z.-M. Zhang, Y.-G. Li, Z.-M. Su, W.-L.
Li, E.-B. Wang, Angew. Chem. 2012, 124, 8109–8113; Angew. Chem. Int.
Ed. 2012, 51, 7985–7989; d) Y. Zhao, J. Zhang, B. Han, J. Song, J. Li, Q.
Wang, Angew. Chem. 2011, 123, 662–665; Angew. Chem. Int. Ed. 2011,
8
08C under vacuum for 24 h to obtain [TMGOH]Br with a modest
yield of 80%.
5
0, 636–639.
The monohydroxy-tethered POM,
a white solid assigned as
[
8] a) A. B. Bourlinos, K. Raman, R. Herrera, Q. Zhang, L. A. Archer, E. P. Gian-
nelis, J. Am. Chem. Soc. 2004, 126, 15358–15359; b) Y. Zhang, Y. Shen, J.
Yuan, D. Han, Z. Wang, Q. Zhang, L. Niu, Angew. Chem. 2006, 118,
5999–6002; Angew. Chem. Int. Ed. 2006, 45, 5867–5870; c) P. G. Rickert,
M. R. Antonio, M. A. Firestone, K. A. Kubatko, T. Szreder, J. F. Wishart,
M. L. Dietz, J. Phys. Chem. B 2007, 111, 4685–4692.
9] a) L. Liu, C. Chen, X. Hu, T. Mohamood, W. Ma, J. Lin, J. Zhao, New J.
Chem. 2008, 32, 283–289; b) S. Wang, W. Liu, Q. Wan, Y. Liu, Green
Chem. 2009, 11, 1589–1594; c) L. Gharnati, O. Walter, U. Arnold, M.
Dçring, Eur. J. Inorg. Chem. 2011, 2756–2762.
10] a) Y. Qiao, Z. Hou, H. Li, Y. Hu, B. Feng, X. Wang, L. Hua, Q. Huang, Green
Chem. 2009, 11, 1955–1960; b) H. Li, Z. Hou, Y. Qiao, B. Feng, Y. Hu, X.
Wang, X. Zhao, Catal. Commun. 2010, 11, 470–475.
11] a) Y. Leng, J. Wang, D. Zhu, X. Ren, H. Ge, L. Shen, Angew. Chem. 2009,
121, 174–177; Angew. Chem. Int. Ed. 2009, 48, 168–171; b) W. Zhang, Y.
Leng, P. Zhao, J. Wang, D. Zhu, J. Huang, Green Chem. 2011, 13, 832–
[
TMGOH] H PW, was synthesized similarly with a yield of 92%.
2.2 0.8
FTIR (KBr): n˜ =3465, 3368, 3257, 2938, 1639, 1618, 1561, 1459,
1
À1
411, 1078, 977, 895, 810, 595, 514 cm ; elemental analysis calcd
(
wt%): C 5.72, H 1.26, N 2.86; found: C 5.70, H 1.30, N 2.89. The hy-
droxy-free [TMG] PW was prepared by neutralization reaction of
H PW O (1 mmol) with TMG (3 mmol) in aqueous solution to
give a white solid, yield 95%. FTIR (KBr): n˜ =3455, 3367, 3252,
2
emental analysis calcd (wt%): C 5.62, H 1.30, N 3.91; found: C 5.68,
H 1.30, N 3.95.
3
[
3
12 40
À1
940, 1617, 559, 1455, 1411, 1079, 978, 895, 810, 596, 520 cm ; el-
[
[
Catalytic tests
The catalytic performance was studied by a model reaction, the
834; c) Y. Leng, J. Wang, D. Zhu, M. Zhang, P. Zhao, Z. Long, J. Huang,
epoxidation of cis-cyclooctene with H O . In a typical run, cis-cyclo-
Green Chem. 2011, 13, 1636–1639; d) P. Zhao, M. Zhang, Y. Wu, J. Wang,
Ind. Eng. Chem. Res. 2012, 51, 6641–6647; e) P. Zhao, Y. Leng, M. Zhang,
J. Wang, Y. Wu, J. Huang, Chem. Commun. 2012, 48, 5721–5723; f) M.
Zhang, P. Zhao, Y. Leng, G. Chen, J. Wang, J. Huang, Chem. Eur. J. 2012,
2
2
octene (10 mmol), methanol (3 mL), and catalyst [TMGDH] H PW
2.3 0.7
(
0.1 g, 0.03 mmol) were stirred in a flask at 608C for several mi-
nutes, and then 30 wt% aqueous H O (5 mmol) was added drop
2
2
1
8, 12773–12782.
by drop. The reaction was continued for 6 h with heating at reflux.
Then the internal standard decane was added and the resulting
mixture was diluted with methanol. The organic phase was collect-
ed and analyzed by gas chromatography, with an instrument (SP-
[
[
12] a) N. M. M. Mateus, L. C. Branco, N. M. T. LourenÅo, C. A. M. Afonso,
Green Chem. 2003, 5, 347–352; b) Y. Gao, S. W. Arritt, B. Twamley, J. M.
Shreeve, Inorg. Chem. 2005, 44, 1704–1712.
13] a) X.-Y. Dou, J.-Q. Wang, Y. Du, E. Wang, L.-N. He, Synlett 2007, 3058–
6
890A) equipped with a flame ionization detector and a capillary
3062; b) Y. Jin, S. Fang, L. Yang, S. Hirano, K. Tachibana, J. Power Sources
column (SE-54, 30 mꢁ0.32 mmꢁ0.3 mm). Conversion (based on
H O )=(mmol cis-cyclooctene converted)/(mmol initial H O )ꢁ
2011, 196, 10658–10666; c) G. V. S. M. Carrera, R. F. M. Frade, J. Aires-de-
Sousa, C. A. M. Afonso, L. C. Branco, Tetrahedron 2010, 66, 8785–8794;
2
2
2
2
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 568

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
d) J. Liu, F. Wang, Z. Li, J. Zhou, J. Chen, C. Xia, Struct. Chem. 2011, 22,
[29] a) Y. Liu, K. Murata, T. Hanaoka, M. Inaba, K. Sakanishi, J. Catal. 2007,
248, 277–287; b) L. Nakka, J. E. Molinari, I. E. Wachs, J. Am. Chem. Soc.
2009, 131, 15544–15554; c) Y. M. A. Yamada, C. K. Jin, Y. Uozumi, Org.
Lett. 2010, 12, 4540–4543.
[30] T. Mallat, A. Baiker, Catal. Sci. Technol. 2011, 1, 1572–1583.
[31] P. G. Jessop, Green Chem. 2011, 13, 1391–1398.
[32] W. Fan, P. Wu, T. Tatsumi, J. Catal. 2008, 256, 62–73.
[33] a) Y. Ishii, K. Yamawaki, T. Ura, H. Yamada, T. Yoshida, M. Ogawa, J. Org.
Chem. 1988, 53, 3581–3593; b) D. C. Duncan, R. C. Chambers, E. Hecht,
C. L. Hill, J. Am. Chem. Soc. 1995, 117, 681–691; c) R. Prabhakar, K. Moro-
kuma, C. L. Hill, D. G. Musaev, Inorg. Chem. 2006, 45, 5703–5709; d) K.
Kamata, M. Kotani, K. Yamaguchi, S. Hikichi, N. Mizuno, Chem. Eur. J.
2007, 13, 639–648; e) K. Kamata, R. Ishimoto, T. Hirano, S. Kuzuya, K.
Uehara, N. Mizuno, Inorg. Chem. 2010, 49, 2471–2478.
1119–1130.
[
14] a) P. J. Cordeiro, T. D. Tilley, ACS Catal. 2011, 1, 455–467; b) F. Xiao, B.
Xie, H. Zhang, L. Wang, X. Meng, W. Zhang, X. Bao, B. Yilmaz, U. Mꢂller,
H. Gies, H. Imai, T. Tatsumi, D. D. Vos, ChemCatChem 2011, 3, 1442–
1
446.
15] K. Kamata, T. Hirano, S. Kuzuya, N. Mizuno, J. Am. Chem. Soc. 2009, 131,
997–7004.
[
6
[
[
[
[
16] L. A. Pꢃrez-Maqueda, E. Matijevic, Chem. Mater. 1998, 10, 1430–1435.
17] X. Yan, P. Zhu, J. Fei, J. Li, Adv. Mater. 2010, 22, 1283–1287.
18] T. Okuhara, T. Nakato, Catal. Surv. Jpn. 1998, 2, 31–44.
19] Y. Jiang, S. Liu, S. Li, J. Miao, J. Zhang, L. Wu, Chem. Commun. 2011, 47,
10287–10289.
[
[
[
20] G. Ranga Rao, T. Rajkumar, B. Varghese, Solid State Sci. 2009, 11, 36–42.
21] M. Misono, Catal. Rev. Sci. Eng. 1987, 29, 269–321.
22] I. Holclajtner-Antunovi c´ , D. Bajuk-Bogdanovi c´ , A. Popa, S. Uskokovi c´ -
Markovi c´ , Inorg. Chim. Acta 2012, 383, 26–32.
[34] N. S. Antonova, J. J. Carbꢅ, U. Kortz, O. A. Kholdeeva, J. M. Poblet, J. Am.
Chem. Soc. 2010, 132, 7488–7497.
[35] A. Berkessel, J. A. Adrio, J. Am. Chem. Soc. 2006, 128, 13412–13420.
[36] M. Ziolek, Catal. Today 2004, 90, 145–150.
[
[
23] J. C. Duhacek, D. C. Duncan, Inorg. Chem. 2007, 46, 7253–7255.
24] D. J. Heldebrant, P. K. Koech, M. Trisha C. Ang, C. Liang, J. E. Rainbolt,
C. R Yonker, P. G. Jessop, Green Chem. 2010, 12, 713–721.
[37] C. L. Hill, J. Mol. Catal. A 2007, 262, 2–6.
[38] D. Demberelnyamba, S. J. Yoon, H. Lee, Chem. Lett. 2004, 33, 560–561.
[39] Z. Wang, Y. Cui, Z. Xu, J. Qu, J. Org. Chem. 2008, 73, 2270–2274.
[40] L. C. Branco, J. N. Rosa, J. J. Moura Ramos, C. A. M. Afonso, Chem. Eur. J.
2002, 8, 3671–3677.
[
[
[
25] F. Bellina, C. Chiappe, M. Lessi, Green Chem. 2012, 14, 148–155.
26] T. Yamase, Chem. Rev. 1998, 98, 307–325.
27] M. Guidotti, C. Pirovano, N. Ravasio, B. Lꢄzaro, J. M. Fraile, J. A. Mayoral,
B. Coq, A. Galarneau, Green Chem. 2009, 11, 1421–1427.
[28] a) O. A. Kholdeeva, N. V. Maksimchuk, G. M. Maksimov, Catal. Today
2
2
010, 157, 107–113; b) Y. Ding, W. Zhao, H. Hua, B. Ma, Green Chem.
008, 10, 910–913; c) L. Hua, Y. Qiao, Y. Yu, W. Zhu, T. Cao, Y. Shi, H. Li,
Received: January 30, 2013
B. Feng, Z. Hou, New J. Chem. 2011, 35, 1836–1841.
Published online on April 26, 2013
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 561 – 569 569