Polyoxovanadometalate-catalyzed selective epoxidation of alkenes with hydrogen peroxide

Article abstract of DOI:10.1002/anie.200500491

The bis(μ-hydroxo)-bridged dioxovanadium site in [γ-1,2-H ₂SiV₂W₁₀O₄₀]⁴ catalyzes the epoxidation of alkenes in the presence of only one equivalent of H ₂O₂ with a high yield of epoxide, high efficiency of H₂O₂ utilization, unusual regioselectivity, and unprecedented diastereoselectivity (see picture). (Chemical Equation Presented)

Full text of DOI:10.1002/anie.200500491

Communications
Catalytic Epoxidation
DOI: 10.1002/anie.200500491
Polyoxovanadometalate-Catalyzed Selective
Epoxidation of Alkenes with Hydrogen
Peroxide**
Yoshinao Nakagawa, Keigo Kamata, Miyuki Kotani,
Kazuya Yamaguchi, and Noritaka Mizuno*
Vanadium is an important element in biology, inorganic
chemistry, and organic syntheses,[1–8] and can heterogeneously
or homogenously catalyze selective oxidation reactions of
alkanes, alkenes, aromatic compounds, alcohols, halides,
etc.[1–4,7–14] While homogenous oxidation catalysis by mono-
and divanadium complexes has extensively been studied, the
oxidation catalysis by bis(m-hydroxo)-bridged divanadium
compounds is scarcely known.[1–8,15,16] In addition, even
monovanadium complexes cannot efficiently catalyze the
epoxidation of alkenes by a green oxidant such as H2O2
because of a contribution from the radical mechanism; for
example, the oxidation of cyclohexene by H2O2 in the
presence of vanadium-based catalysts proceeds mainly at
the allylic position.[9–14] Moreover, the efficiency of H2O2
utilization is intrinsically low.[9–14] Therefore the vanadium-
catalyzed, H2O2-based epoxidation of alkenes with high
selectivity to the epoxide and high efficiency of H2O2
utilization is currently not possible, although it is well
known that other transition metals such as tungsten, man-
ganese, and iron show high activity and selectivity.[17–20]
[*] Dr. Y. Nakagawa, Dr. K. Yamaguchi, Prof. Dr. N. Mizuno
Department of Applied Chemistry
School of Engineering
The University of Tokyo
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656 (Japan)
Fax: (+81)3-5841-7220
E-mail: tmizuno@mail.ecc.u-tokyo.ac.jp
K. Kamata, M. Kotani, Dr. K. Yamaguchi, Prof. Dr. N. Mizuno
Core Research for Evolutional Science and Technology (CREST)
Japan Science and Technology Agency (JST)
4-1-8 Honcho, Kawaguchi, Saitama 332-0012 (Japan)
[**] This work was supported by the CRESTprogram of JSTand a Grant-
in-Aid for Scientific Research from the Ministry of Education,
Culture, Science, Sports, and Technology of Japan.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5136
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 5136 –5141

Angewandte
Chemie
The catalytic function of polyoxometalates has attracted
much attention because the ability to design catalytically
active sites means that their redox and acidic properties can
be controlled at the atomic or molecular levels.[21–28] Recently,
various efficient catalytic systems for H2O2-based epoxidation
by polyoxometalates have been developed with peroxotung-
states or -molybdates as catalyst precursors or with transition-
metal-substituted polyoxometalates. However, these require
either an excess of H2O2 with respect to the alkene or an
excess of alkene with respect to H2O2 to attain a high yield of
epoxide or high efficiency of H2O2 utilization.[29] Herein, we
report that [g-1,2-H2SiV2W10O40]4ꢀ (1), which has a {VO-(m-
OH)2-VO} core, can catalyze the epoxidation of alkenes in the
presence of only one equivalent of H2O2 with a high yield of
epoxide and high efficiency of H2O2 use. Notably, this system
shows unique stereospecificity, diastereoselectivity, and regio-
selectivity that are quite different from those reported for the
epoxidation systems based on [g-SiW10O34(H2O)2]4ꢀ.[30,31]
Furthermore, we have investigated the active site of 1 and
the active species for the present epoxidation on the basis of
various physicochemical characterizations.
reaction conditions, the amounts of allylic oxidation products
and glycols produced by hydrolysis were negligible in all
cases. The epoxidation did not proceed at all in the absence of
catalyst, and no epoxidation was observed with tert-butylhy-
droperoxide (TBHP) as the oxidant. Nonactivated aliphatic
terminal C3–C10 alkenes, including propene, could be trans-
formed into the corresponding epoxides with ꢁ 99% selec-
tivity and ꢁ 87% efficiency of H2O2 utilization. A larger-scale
reaction (50-fold scale-up) for 1-octene showed the same
results as the small-scale experiments (99% selectivity to 1,2-
epoxyoctane, 93% yield determined by GC, and 86% yield of
isolated product). The reaction rate increased with an
increase in the reaction temperature, and the TOF reached
as high as 38 hꢀ1 at 313 K whilst maintaining the selectivity for
1,2-epoxyoctane as high as > 99%, although the efficiency of
H2O2 utilization decreased slightly from 93% (293 K) to 85%
(313 K). Acid-sensitive styrene was epoxidized to give styrene
oxide as the sole product. Also, the catalytic epoxidation of
cyclic alkenes such as cyclohexene and cyclooctene proceeded
efficiently to afford the corresponding epoxides in high yields.
The epoxidation of cis- and trans-2-octene gave cis-2,3-
epoxyoctane (90% yield; Table 1) and trans-2,3-epoxyoctane
First, we examined the scope of the present epoxidation
with the tetra-n-butylammonium salt of 1 (TBA-1), various
kinds of alkenes, and with only one equivalent of H2O2
(except for gaseous alkenes; Table 1). Under the present
=
(6% yield), respectively; the configuration around the C C
moieties is retained in these epoxides. Moreover, for the
competitive epoxidation of cis- and trans-2-octenes (100 mmol
each, the other conditions were the same as in Table 1), the
initial rates for the epoxidation were 0.32 and 0.001 mm minꢀ1,
respectively. The ratio of the formation rate of cis-2,3-
epoxyoctane to that of the trans isomer is more than 3 ” 102,
which is much larger than the ratios (1.3–11.5) reported for
other stereospecific epoxidation systems (see Supporting
Information). For competitive epoxidation of cis- and trans-
2-octenes, the yields of cis-2,3-epoxyoctane and the trans
isomer (based on H2O2), even after 24h, were 91 and 2%,
respectively. The epoxidation of 3-substituted cyclohexenes,
such as 3-methyl-1-cyclohexene and 2-cyclohexen-1-ol,
showed an unusual diastereoselectivity: the corresponding
epoxides are formed highly diastereoselectively with the
oxirane ring trans to the substituents (anti configuration;
Table 1 and Tables S2 and S3 in the Supporting Information).
Further, the more accessible, but less nucleophilic double
bonds in nonconjugated dienes such as trans-1,4-hexadiene,
(R)-(+)-limonene, 7-methyl-1,6-octadiene, and 1-methyl-1,4-
cyclohexadiene were epoxidized in high yields in a highly
regioselective manner, as shown in Table 2. The values of
[less-substituted epoxide]/[total epoxides] (ꢁ 0.88) are much
higher than those reported for the other epoxidation systems
(Tables S4–S7 in the Supporting Information). As far as we
know, such unique stereospecificity, diastereoselectivity, and
regioselectivity have never been reported for 1.
Table 1: Epoxidation of various monoalkenes catalyzed by TBA-1.[a]
Substrate
Product
Yield Selectivity Eox
[%]
[%]
[%][b]
87
91
92
93
99
99
99
99
87
91
92
93
99
(only cis)
90
90
88
93
98
99
90
93
88
99
99
88
90
93
92
93
99
97
91
87
(syn/anti 91
=5/95)
95
(syn/anti 91
=12/88)
The relative reactivity (rate) of a series of C8 alkenes
decreased in the order of 2-methyl-1-heptene (1.1) > 1-
octene (1.0 taken as unity) > cis-2-octene (0.87) > 2-methyl-
2-heptene (0.03) > trans-2-octene (< 0.01) (see Table S8 in
the Supporting Information). This order is not consistent with
[a] Reaction conditions: alkene (33.3 mm), TBA-1 (1.67 mm), H2O2 (30%
aq., 33.3 mm), and CH3CN/tBuOH (1.5/1.5 mL) at 293 K for 24 h. Yields
and selectivities were determined by gas chromatography or 1H NMR
spectroscopy using an internal standard technique and are based on
alkenes. [b] Eox(H2O2 efficiency)=[products(mol)/consumed H2O2 -
(mol)]”100. [c] Propene pressure 6 atm. [d] 1-Butene pressure 3 atm.
=
that of the p(C C) HOMO energies of C8 alkenes because
they decrease with an increase in the number of alkyl
substituents in alkenes, as reported previously.[30] Therefore,
=
the inconsistency between the ordering of the p(C C)
Angew. Chem. Int. Ed. 2005, 44, 5136 –5141
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5137

Communications
Table 2: Epoxidation of various nonconjugated dienes catalyzed by TBA-
also suggests the maintenance of the g-Keggin framework of
1.
1.[a]
Yield [%] [less-substi- Eox [%][b]
tuted epox-
All these facts strongly indicate that the {VO-(m-OH)2-
VO} core is preserved in the cluster framework of 1 during the
catalytic process and that tungstate compounds are not the
true catalytically active species in the present epoxidation.
The fact that the regioselectivities for nonconjugated dienes
with 1 are different from those with [g-SiW10O34(H2O)2]4ꢀ and
[W2O3(O2)4(H2O)2]2ꢀ also support this idea.
Next, the reactivity of 1 with H2O2 was investigated. The
cold-spray ionization mass spectrum (CSI-MS) of TBA-1 in
1,2-C2H4Cl2 shows the most intense parent-ion peak (centered
at m/z 3337.9) with an isotopic distribution (Figure 1a and
Substrate
Product
ide]/[total
epoxides]
91
89
>0.99
91
91
0.99
1
77
5
0.93
0.88
87
81
4
71
5
[a],[b] See footnotes to Table 1.
HOMO energies and the reactivities should be caused by the
steric constraints of the active site of 1, that is, an active
oxygen species formed by the reaction of 1 with H2O2 is
embedded in the polyoxometalate framework and thus a
=
sterically hindered C C double bond cannot easily approach
the site.
The mono- and trivanadium-substituted compounds [a-
SiVW11O40]5ꢀ and [a-1,2,3-SiV3W9O40]7ꢀ are inactive for the
epoxidation under the present conditions, which suggests that
ꢀ ꢀ
=
the V O W and V O centers are not the active sites. The
fully occupied silicotungstate [g-SiW12O40]4ꢀ, which has the
same structure as 1, is also inactive, which suggests that the
tungsten atoms in 1 are not the active sites. The regioselec-
tivity of the catalytic epoxidation of nonconjugated dienes by
1 is quite different from that by [g-SiW10O34(H2O)2]4ꢀ
(Tables S4–S7 in the Supporting Information), which shows
that the catalytically active site (species) of 1 is different from
that of [g-SiW10O34(H2O)2]4ꢀ.
In addition, we confirmed by in situ IR and NMR
measurements that tungstate compounds such as [g-SiW10O34-
(H2O)2]4ꢀ and [W2O3(O2)4(H2O)2]2ꢀ are not formed during
the catalysis. The 51V NMR spectrum of the recovered
Figure 1. CSI-MS (anion mode, m/z 3300–3400) of a) TBA-1 (0.14 mm)
in 1,2-C2H4Cl2 and c) TBA-1 (0.14 mm) in 1,2-C2H4Cl2 treated with
350 equiv of H2O2 (96% aqueous solution) with respect to TBA-1
(253 K, 2 h). The lines in (b) are the pattern calculated for [(TBA)3H2-
SiV2W10O40]ꢀ and the lines in (d) are the pattern calculated for a mix-
ture of [(TBA)3H2SiV2W10O40]ꢀ (22%) and [(TBA)3HSiV2W10O39(OOH)]ꢀ
(78%). The small increase in the content of [(TBA)3H2SiV2W10O40]ꢀ in
comparison with those of Figures 2–4 is probably caused by a slight
increase in the temperature at the inlet of the injection port of the
spectrometer.
catalyst in CD3CN shows a line at d = ꢀ564ppm ( Dn1/2
=
Supporting Information) that agrees with the pattern calcu-
lated for [(TBA)3H2SiV2W10O40]ꢀ (Figure 1b and Supporting
Information). Upon the addition of 350 equiv of H2O2, a new
peak centered at m/z (3337.9 + 16) appeared (Figure 1c)
125 Hz) and the 183W NMR spectrum in CD3CN shows
three lines at d = ꢀ84( Dn1/2 = 9.9 Hz), ꢀ97 (Dn1/2 = 2.3 Hz),
and ꢀ131 ppm (Dn1/2 = 1.6 Hz) with an integrated intensity
ratio of 2:1:2, in agreement with the g-Keggin framework.
Signals due to tungstate compounds such as [a-SiW12O40]4ꢀ,
which
agrees
with
the
pattern
calculated
for
[(TBA)3HSiV2W10O39(OOH)]ꢀ (Figure 1d).
[g-SiW10O34(H2O)2]4ꢀ,
[W2O3(O2)4(H2O)2]2ꢀ
,
and
The H NMR spectrum of 1 (2.5 mm) in 1,2-C2D4Cl2 at
253 K shows a signal at d = 4.80 ppm due to the hydroxy
proton of the {VO-(m-OH)2-VO} core in 1. Upon the addition
of 20 equiv of H2O2, two signals appear at d = 5.10 and
1
[HnWO2(O2)2](2ꢀn)ꢀ were not observed. The IR and UV/Vis
spectra of the catalyst recovered after the completion of the
epoxidation were the same as those of the fresh one, which
5138
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 5136 –5141

Angewandte
Chemie
9.45 ppm with an intensity ratio of 1:1, and the intensity of the
signal at d = 4.80 ppm has greatly decreased (by (93 ꢂ 1)%)
after 2 h, although the sum of the three signal intensities
remains almost constant (Figure 2). A similar lower-field shift
Figure 3. 51V NMR spectra of the same samples as for Figure 2.
Figure 2. 1H NMR spectra of a) TBA-1 (2.5 mm) in 1,2-C2D4Cl2 and
b) TBA-1 (2.5 mm) in 1,2-C2D4Cl2 treated with 20 equiv of H2O2 (95%
aqueous solution) with respect to TBA-1 at 253 K for 2 h. The asterisks
indicate the signals of H2O2 in the organic and aqueous phases.
of the hydroxy proton signal from d = 5.10 to d = 5.18 ppm has
been observed upon the formation of [g-1,2-SiV2W10O38(m-
OH)(m-OCH3)]4ꢀ in CD3CN.[32] The chemical shift of d =
9.45 ppm is close to those of other hydroperoxide species
such as H2O2 (d = 8.74ppm), tBuOOH (d = 8.84ppm), and
PhC(CH3)2OOH (d = 8.95 ppm).
One new 51V NMR signal appears at d = ꢀ530 ppm in 1,2-
C2D4Cl2 upon the addition of H2O2, and the intensity of the
signal of 1 at d = ꢀ562 ppm has greatly decreased (by (93 ꢂ
1)%) after 2 h, although, again, the sum of the two signal
intensities remains almost constant (Figure 3). Six new
183W NMR signals appear at d = ꢀ79, ꢀ83, ꢀ92, ꢀ104,
ꢀ127, and ꢀ132 ppm with an intensity ratio of 2:2:1:1:2:2,
in 1,2-C2D4Cl2 upon the addition of 15 equiv of H2O2, and the
signals at d = ꢀ81, ꢀ95, and ꢀ128 ppm due to 1 have almost
completely disappeared after 2 h, although the sum of the
nine signal intensities remains almost constant (Figure 4). A
quantitative analysis of the 1H, 51V, and 183W NMR data gives
an atomic ratio of H(d=5.10 ppm):H(9.45):V(ꢀ530):
W(ꢀ79):W(ꢀ83):W(ꢀ92):W(ꢀ104):W(ꢀ127):W(ꢀ132) of
1:1:2:2:2:1:1:2:2.
Figure 4. 183W NMR spectra of a) TDA-1 (0.1m) in 1,2-C2D4Cl2 and
(b) TDA-1 (0.1m) in 1,2-C2D4Cl2 treated with 15 equiv of H2O2 (95%
aqueous solution) with respect to TDA-1 at 253 K for 2 h.
TDA=n-decylammonium.
The 183W NMR spectral pattern depends on the structure
and symmetry of the anion. The anion 1 has C2v symmetry and
shows three signals with intensities of 2:1:2. The reaction of 1
with H2O2 lowers or does not change the symmetry of the
anion, and there are five possibilities for the symmetry of the
resulting compound: 1) C2, 2) Cs with a symmetry plane
containing two vanadium atoms, 3) Cs with a symmetry plane
Angew. Chem. Int. Ed. 2005, 44, 5136 –5141
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5139

Communications
ꢀ ꢀ
containing two V O V bridging oxygen atoms, 4) C1, and
Experimental Section
5) C2v. After the treatment of 1 with H2O2, six 183W NMR
signals are observed with an intensity ratio of 2:2:1:1:2:2; only
case (3) can explain the observed 183W NMR spectrum.[33] The
1H NMR spectrum shows two signals at d = 5.14(1H) and
9.45 ppm (1H). These two protons must exist in the symmetry
plane, which favors the Cs symmetry. The 51V NMR spectrum
shows one signal. Therefore, the NMR data indicate the
formation of a {VO-(m-OH)(m-OOH)-VO} species.
An aqueous solution of [g-1,2-H2SiV2W10O40]4ꢀ was prepared accord-
ing to reference [36], and the anion was isolated as the tetra-n-
butylammonium salt (TBA-1): K8[g-SiW10O36]·12H2O[37] (8 g,
2.7 mmol) was quickly dissolved in 1m HCl (28 mL), NaVO3 (0.5m,
11 mL, 5.5 mmol) was added, and the mixture was gently stirred for
5 min. The solution was filtered and [(n-C4H9)4N]Br (8 g, 25 mmol)
added in a single step. The resulting yellow precipitate was collected
by filtration and then washed with a large amount of water (300 mL).
The crude product was purified twice by precipitation (addition of 1 L
of H2O to an acetonitrile solution of TBA-1 (50 mL)). Analytically
pure TBA-1 was obtained as a pale-yellow powder. Yield: 7.43 g
(76%). Anal. calcd. for [(C4H9)4N]4[g-1,2-H2SiV2W10O40]·H2O: C
21.4, H 4.15, N 1.56, Si 0.78, V 2.83, W 51.1, H2O 0.50; found: C 21.4,
H 3.91, N 1.59, Si 0.79, V 2.88, W 51.2, H2O 0.50. 51V NMR (CD3CN):
An intense signal at d = ꢀ563.6 ppm (Dn1/2 = 133 Hz) attributed to
two equivalent vanadium atoms is observed. 183W NMR (CD3CN):
d = ꢀ82.2 (Dn1/2 = 9.6 Hz), ꢀ95.6 (Dn1/2 = 2.5 Hz), and ꢀ129.7 ppm
(Dn1/2 = 2.9 Hz) with an integrated intensity ratio of 2:1:2. 29Si NMR
(CD3CN): d = ꢀ84.0 ppm (Dn1/2 = 2.0 Hz). The UV/Vis spectrum (in
CH3CN) shows shoulder bands at 240 (e = 36000 mꢀ1 cmꢀ1), 285
(24000), and 350 nm (5900), characteristic of the g-Keggin struc-
ture.[38,39] IR (KBr): n˜ = 1151, 1106, 1057, 1004, 995, 966, 915, 904, 875,
The chemical shifts of the 51V and 183W NMR signals are
also in agreement with this idea: the oxo!peroxo trans-
formation for complexes of 51V, 95Mo, and 183W leads to
significant upfield shifts in the NMR signals (see Table S11 in
the Supporting Information). On the other hand, the oxo!
hydroperoxo transformation from HxMoO2(O2)2(2ꢀx)ꢀ (0 ꢃ x ꢃ
2ꢀ
1) to [MoO(O2)2OOH]2 produces a 15–30 ppm downfield
shift in the 95Mo signals;[34] no 51V NMR data have been
reported for the oxo!hydroperoxo transformation as far as
we know. In the present case, a downfield shift of 32 ppm is
observed for the 51V NMR signal after the treatment of 1 with
H2O2. Such a downfield shift is in accordance with the oxo!
hydroperoxo transformation. These NMR and CSI-MS data
strongly suggest that the {VO-(m-OH)2-VO} core in 1 reacts
with H2O2 to form a {VO-(m-OH)(m-OOH)-VO} species.
The methanol monoester of 1 (2) was easily formed by the
dehydrative condensation between the hydroxy group in 1
and methanol (equilibrium constant 75).[32] No diester was
formed even in the presence of an excess of methanol. The
anionic cluster 2 retains the g-Keggin structure with the two
vanadium atoms bridged by one protonated oxygen atom and
one methoxy group.[32] Therefore, the dehydrative condensa-
tion between 1 and H2O2 to form the hydroperoxo complex
should proceed as reported for bis(m-hydroxo) complexes of
the first row transition metals.[35] When one proton in the
{VO-(m-OH)2-VO} site in 1 was removed by titration with
1 equivalent of nBu4NOH with respect to 1, neither ester-
ification of 1 with methanol nor its epoxidation with H2O2
proceeded. All these results indicate that the {VO-(m-OH)2-
VO} core in 1 is the active site that forms {VO-(m-OH)(m-
OOH)-VO} species for the epoxidation of olefins with H2O2,
according to Equation (1) with R = OH, CH3.
840, 790, 691, 550, 519, 482, 457, 405 cmꢀ1
.
The epoxidation of gaseous substrates (propene and 1-butene)
was carried out in a Teflon-coated autoclave; a glass-tube reactor was
used for the other substrates. The epoxidation was carried out as
follows: catalyst (1.67 mm), CH3CN/tBuOH (1.5/1.5 mL), H2O2 (30%
aq., 33.3 mm), and substrate (propene, 6 atm; 1-butene, 1 atm; others,
33.3 mm) were charged in the reaction vessel. The reaction was
carried out at (293 ꢂ 0.2) K. The reaction solution was sampled
periodically and analyzed by GC in combination with mass spec-
trometry. The products were identified by comparing their mass and
NMR spectra with those of authentic samples. The carbon balance in
each experiment was in the range of 95–100%. The amount of H2O2
remaining after the reaction was analyzed by Ce4+/Ce3+ titration.[40]
After the reaction, the catalyst was recovered by evaporating the
reaction mixture to dryness, followed by washing with tBuOH and n-
hexane. TOF values were estimated from the initial rates below 10%
conversion of H2O2.
Received: February 9, 2005
Revised: May 24, 2005
Published online: July 11, 2005
Keywords: alkenes · epoxidation · hydrogen peroxide ·
.
polyoxometalates · vanadium
½g-1,2-H2SiV2W10O40Š4ꢀ þ ROH
ð1Þ
! ½g-1,2-SiV2W10O38ðm-OHÞðm-ORފ4ꢀ þ H2O
[1] T. Hirao, Coord. Chem. Rev. 2003, 237, 1.
[2] T. Hirao, Chem. Rev. 1997, 97, 2707.
As has been reported previously,[32] the esterification of
the {VO-(m-OH)2-VO} core in 1 with alcohols is sterically
controlled: the esters of primary alcohols with a bulky
substituent, such as neopentyl alcohol, or of secondary and
tertiary alcohols are hardly formed (equilibrium constant
< 0.02) because of the steric repulsion from the polyoxo-
metalate framework. For the same reason, no reaction of 1
with TBHP occurs. Thus, the hydroperoxo group in the
possible intermediate [g-1,2-SiV2W10O38(m-OH)(m-OOH)]4ꢀ
may also be sterically hindered, which could explain the
unique stereospecificity, diastereoselectivity, and regioselec-
tivity in the epoxidation catalysis (Figure S3 in the Supporting
Information).
[3] A. E. Shilov, G. B. Shulꢀpin, Chem. Rev. 1997, 97, 2879.
[4] Thematic issue on vanadium-based catalysts for the selective
oxidation of hydrocarbons and their derivatives: Appl. Catal. A
1997, 157, 1 – 4 25.
[5] H. Sigel, A. Sigel, Vanadium and Its Role in Life, Marcel Dekker,
New York, 1995.
[6] F. Jiang, O. P. Anderson, S. M. Miller, J. Chen, M. Mahroof-
Tahir, D. C. Crans, Inorg. Chem. 1998, 37, 5439.
[7] A. Butler, M. J. Clague, G. E. Meister, Chem. Rev. 1994, 94, 625.
[8] A. G. J. Ligtenbarg, R. Hage, B. L. Feringa, Coord. Chem. Rev.
2003, 237, 89.
[9] U. Schuchardt, M. C. Guerreiro, G. B. Shulꢀpin, Russ. Chem.
Bull. 1998, 107, 247.
[10] O. Bortolini, F. Di Furia, P. Scrimin, G. Modena, J. Mol. Catal.
1980, 7, 59.
5140
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 5136 –5141

Angewandte
Chemie
[11] E. J. Eisenbraun, A. R. Bader, J. W. Polacheck, E. Reif, J. Org.
Chem. 1963, 28, 2057.
[12] R. Neumann, M. Levin-Elad, Appl. Catal. A 1995, 122, 85.
[13] D. Wei, W. Chuei, G. L. Haller, Catal. Today 1999, 51, 501.
[14] G. Bellussi, M. S. Rigutto, Stud. Surf. Sci. Catal. 1994, 85, 177.
[15] A. M. Khenkin, L. Weiner, Y. Wang, R. Neumann, J. Am. Chem.
Soc. 2001, 123, 8531.
[16] R. D. Gall, M. Faraj, C. L. Hill, Inorg. Chem. 1994, 33, 5015.
[17] B. S. Lane, K. Burgess, Chem. Rev. 2003, 103, 2457.
[18] G. Grigoropoulou, J. H. Clark, J. A. Elings, Green Chem. 2003, 5,
1.
[19] R. Noyori, M. Aoki, K. Sato, Chem. Commun. 2003, 1977.
[20] I. W. C. E. Arends, R. A. Sheldon, Top. Catal. 2002, 19, 133.
[21] M. T. Pope in Comprehensive Coordination Chemistry II, Vol. 4
(Eds.: J. A. McCleverty, T. J. Meyer), Elsevier, Amsterdam,
2003, p. 635 – 678.
[22] C. L. Hill in Comprehensive Coordination Chemistry II, Vol. 4
(Eds.: J. A. McCleverty, T. J. Meyer), Elsevier, Amsterdam,
2003, p. 679 – 759.
[23] I. V. Kozhevnikov, Catalysis by Polyoxometalates, Wiley, Chi-
chester, UK, 2002.
[24] I. V. Kozhevnikov, Chem. Rev. 1998, 98, 171.
[25] N. Mizuno, M. Misono, Chem. Rev. 1998, 98, 199.
[26] R. Neumann, Prog. Inorg. Chem. 1998, 47, 317.
[27] T. Okuhara, N. Mizuno, M. Misono, Adv. Catal. 1996, 41, 113.
[28] C. L. Hill, C. M. Prosser-McCartha, Coord. Chem. Rev. 1995,
143, 407.
[29] See Table S10 in the Supporting Information.
[30] K. Kamata, Y. Nakagawa, K. Yamaguchi, N. Mizuno, J. Catal.
2004, 224, 224.
[31] K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S. Hikichi,
N. Mizuno, Science 2003, 300, 964.
[32] Y. Nakagawa, K. Uehara, N. Mizuno, Inorg. Chem. 2005, 44, 14 .
[33] In the cases (1), (2), (4), and (5), the 183W NMR spectra should
show five lines (intensity ratio 1:1:1:1:1), five lines (1:1:1:1:1), ten
lines (1:1:1:1:1:1:1:1:1:1), and three lines (2:2:1), respectively.
[34] V. Nardello, J. Marko, G. Vermeersch, J. M. Aubry, Inorg. Chem.
1995, 34, 4950.
[35] M. Akita, S. Hikichi, Bull. Chem. Soc. Jpn. 2002, 75, 1657.
[36] J. Canny, R. Thouvenot, A. Tꢁzꢁ, G. Hervꢁ, M. Leparulo-Loftus,
M. T. Pope, Inorg. Chem. 1991, 30, 976.
[37] A. Tꢁzꢁ, G. Hervꢁ, Inorg. Synth. 1990, 27, 85.
[38] A. Tꢁzꢁ, J. Canny, L. Gurdan, R. Thouvenot, G. Hervꢁ, Inorg.
Chem. 1996, 25, 1001.
[39] N. Mizuno, C. Nozaki, I. Kiyoto, M. Misono, J. Am. Chem. Soc.
1998, 120, 9267.
[40] A. I. Vogel, A Textbook of Quantitative Inorganic Analysis
Including Elementary Instrumental Analysis, Longman, New
York, 1978.
Angew. Chem. Int. Ed. 2005, 44, 5136 –5141
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5141