Synthesis, structural characterization, and oxidation catalysis of a diniobium-substituted silicodecatungstate

Article abstract of DOI:10.1246/cl.150213

A novel diniobium-substituted silicodecatungstate, [γ-HSiW₁₀O₃₈Nb₂(η²-O₂)₂]^5- (I), was successfully synthesized by the reaction of [HNb₆O₁₉]^7- and [γ-SiW₁₀O₃₆]^8- in the presence of H₂O₂ under acidic conditions. In the presence of a suitable proton source, e.g., HClO₄, I could act as a stable homogeneous catalyst for H₂O₂-based oxidation.

Full text of DOI:10.1246/cl.150213

Received: March 10, 2015 | Accepted: April 3, 2015 | Web Released: April 9, 2015
CL-150213
Synthesis, Structural Characterization, and Oxidation Catalysis
of a Diniobium-substituted Silicodecatungstate
³
Naoto Satake, Tomohisa Hirano, Keigo Kamata, Kosuke Suzuki, Kazuya Yamaguchi, and Noritaka Mizuno*
Department of Applied Chemistry, School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656
(E-mail: tmizuno@mail.ecc.u-tokyo.ac.jp)
A
novel diniobium-substituted silicodecatungstate, [γ-
were synthesized by the reaction of K7[HNb6O19]¢13H2O, K8[γ-
SiW10O36]¢12H2O, and the corresponding alkylammonium
bromide in the presence of H2O2 under acidic conditions
(see the Supporting Information). The tetramethylammonium
(TMA) salt of I (TMA5I) was also synthesized by the cation-
exchange reaction of TBA5I with TMANO3. The IR spectra
of TMA5I, TBA5I, and TPeA5I showed bands around 830,
HSiW10O38Nb2(η2-O2)2]5¹ (I), was successfully synthesized by
the reaction of [HNb6O19]7¹ and [γ-SiW10O36]8¹ in the presence
of H2O2 under acidic conditions. In the presence of a suitable
proton source, e.g., HClO4, I could act as a stable homogeneous
catalyst for H2O2-based oxidation.
¹1
600, and 550 cm assignable to v(O-O), vasym(Nb(O2)), and
Polyoxometalates (POMs) are early transition-metal (V, Nb,
Ta, Mo, W, etc.) oxygen cluster anions with discrete and
versatile structures and have attracted much attention in various
fields of science such as catalysis, medicine, and advanced
material science.1 In particular, POMs are promising oxidation
catalysts owing to the following advantages: (i) Catalytically
active sites can precisely be created at atomic and/or molecular
levels and (ii) they are relatively stable under thermal and
oxidative conditions.2 Hydrogen peroxide (H2O2) is an ideal
oxidant because of its high content of active oxygen species
and coproduction of only water. Therefore, a large number of
H2O2-based oxidation systems with POM-based catalysts such
as peroxometalates, lacunary POMs, and transition-metal-sub-
stituted POMs have been developed to date.3
vsym(Nb(O2)), respectively, and these band positions are close to
those of typical peroxoniobates (Figure S1).6-8
Single crystals of TMA5I suitable for X-ray crystallographic
analysis were successfully obtained from a mixed solvent of
acetonitrile and water containing crude TMA5I with vapor
diffusion of diethyl ether. The crystallographic data are summa-
rized in Table S1, and the ORTEP representation of the anionic
part of TMA5I is shown in Figure 1. The anionic part of TMA5I
was a monomer of a diniobium-substituted γ-Keggin POM.
One terminal η2-peroxo group was coordinated to each of the
two niobium atoms, and the O-O bond lengths (1.46-1.47 ¡)
were comparable to those of typical peroxoniobates (1.43-
1.51 ¡) (Table S2).6-8 The elemental analysis and X-ray crys-
tallographic analysis data revealed the existence of five TMA
cations per anion. The bond valence sum (BVS) values of
niobium (5.27-5.32), tungsten (6.04-6.20), and silicon (3.75)
indicated that the respective valences were +5, +6, and +4.
Therefore, one proton is probably associated with I. The BVS
value of O38 (1.12) was significantly lower than those of other
oxygen atoms (1.72-2.09), showing that O38 was possibly
protonated. To the best of our knowledge, this is the first
example of a structurally characterized diniobium-substituted
γ-Keggin silicodecatungstate.
It is known that d0-transition-metal-substituted POMs can
efficiently catalyze the H2O2-based oxidation of various kinds of
organic substrates.4,5 In particular, the activation of H2O2 by
titanium-4 and vanadium-substituted POMs,5 and their oxidation
catalysis have extensively been investigated. On the other hand,
there are only a few reports on H2O2-based oxidation catalysis
by niobium-substituted POMs.6 While allylic alcohols are
selectively oxidized to the corresponding triols with H2O2
in the presence of triniobium-substituted POMs such as
[H2Si2W18Nb6O77]6¹ and A,β-[PW9O37(NbO2)3]6¹, these POMs
are not active for the epoxidation of simple alkenes.6 In addition,
these POMs are decomposed into peroxotungstate fragments to
some extent because of the presence of excess amounts of H2O2,
which are the truly catalytically active species for the oxidation
of allylic alcohols.6 Therefore, successful examples of POMs
with well-defined niobium centers, which can activate H2O2 and
catalyze the oxidation of various substrates, including simple
alkenes, have not yet been reported.
Herein, we report the successful synthesis and structural
characterization of a novel diniobium-substituted silicodeca-
tungstate, [γ-HSiW10O38Nb2(η2-O2)2]5¹ (I). In the presence of a
suitable proton source, e.g., HClO4, I could act as a stable
homogeneous catalyst for the H2O2-based oxidation of several
organic substrates, e.g., cyclooctene, thioanisole, 1-phenyl-
ethanol, and allyl alcohol. The reaction mechanism, including
the effect of protons on the formation of active species, was also
investigated.
The positive-ion cold-spray ionization mass (CSI-MS)
spectrum of TMA5I in CD3CN exhibited two sets of signals
centered at m/z 3170 and 1622, with isotopic distributions
O40
O42
O41
O39
O1
O38
Nb1
O2
Nb2
O4
O3
O37
O7
O10
O9
O13
O12
W2
O5
O18
O16
O6
W6
W4
O15
W3
O25
O26
O11
Si1
W1
O8
O36
O35
O34
O20
O22
O14
O19
O28
W8
W10
O21
O32
W5
O17
O30
O29
O27
W7
W9
O24
O33
O23
O31
The tetra-n-butylammonium (TBA) and tetra-n-pentylam-
monium (TPeA) salts of I (TBA5I and TPeA5I, respectively)
Figure 1. ORTEP drawing of the anion part of TMA5I drawn at
50% probability level.
Chem. Lett. 2015, 44, 899–901 | doi:10.1246/cl.150213
© 2015 The Chemical Society of Japan | 899

Table 1. Effect of proton on the TBA5I-catalyzed epoxidation of 1
with H2O2
(b)
(d)
(a)
(c)
a
OH
OH
O
O
O
+
+
+
OH
1
2
3
4
5
R0
/mM min
Entry HClO4/equiv
Yield/% Selectivity to 2/%
¹1
¹3
¹2
¹2
¹2
1
2
3
4
None
(1.7 « 0.9) © 10
(1.3 « 0.2) © 10
(5.4 « 0.7) © 10
(5.4 « 0.4) © 10
9
48
60
67
>99
98
94
1
2
3
91
–80 –82 –84 –86 –88 –90 –70 –90 –110 –130 –150 –170
chemical shift / ppm chemical shift / ppm
aReaction conditions: TBA5I (4 ¯mol), 1 (1.0 mmol), 30% H2O2
(0.1 mmol), HClO4 (0-3 equiv with respect to TBA5I), CH3CN
(3 mL), 283 K, 24 h. Yield and selectivity were determined by GC
analysis using chlorobenzene as the internal standard. Yield (%) =
total products (mol)/initial H2O2 (mol) © 100. Selectivity (%) =
desired product (mol)/total products (mol) © 100.
Figure 2. (a) 29Si and (b) 183W NMR spectra of TPeA5I. (c) 29Si and
(d) 183W NMR spectra of TPeA5I in the presence of H2O2 and two
equiv of HClO4. Conditions for (a) and (c): TPeA5I (0.03 M), 97%
H2O2 (0 or 2.4 M), HClO4 (0 or 0.06 M), CD3CN, 233 K. Conditions
for (b) and (d): TPeA5I (0.1 M), 97% H2O2 (0 or 8.0 M), HClO4 (0 or
0.2 M), CD3CN, 233 K (see also Figure S3).
For early transition-metal complexes, protons play an
important role in the formation of active species for epoxidation
with H2O2.10 In order to confirm the formation of the active
species, the NMR (29Si and 183W) and CSI-MS spectra of
TPeA5I were measured in the presence of both H2O2 and protons
(2 equiv of HClO4).9 The 29Si NMR spectrum showed one signal
at ¹84.9 ppm (Figure 2c), showing the formation of a single
species. Upon addition of H2O2 and HClO4, six signals appeared
at ¹109.8, ¹114.0, ¹120.5, ¹138.6, ¹139.6, and ¹150.4 ppm
with the respective intensity ratio of 2:1:2:2:1:2, and the
chemical shifts were significantly different from those of I
(Figure 2d). The 2:1:2:2:1:2 ratio suggested that the species
also had Cs symmetry. The CSI-MS spectrum of TPeA5I in the
presence of H2O2 and 2 equiv of HClO4 exhibited a set of
intense signals centered at m/z 4400 assignable to a triperoxo
species, [TPeA5II¢2CH3CN¢HClO4]+ (II = [SiW10O36Nb2-
(O2)3]4¹) (Figure S6a). Upon addition of 1 to the solution, the
signals almost disappeared and new signals centered at m/z
4384 assignable to [TPeA5SiW10O37Nb2(O2)2¢2CH3CN¢
HClO4]+ appeared (Figure S6b). These results show that II
with Cs symmetry is the possible active species for epoxidation.
Kinetic studies on the epoxidation of 1 with H2O2 catalyzed
by TBA5I in the presence of 2 equiv of HClO4 showed the first-
order dependence of the reaction rates on the concentration of
TBA5I (0.67-2.67 mM, Figure S7a), while it has been reported
that fractional order dependence is observed when using
[H2Si2W18Nb6O77]6¹ and A,β-[PW9O37(NbO2)3]6¹.6 In addition,
the NMR and CSI-MS signals assignable to peroxotungstate
fragments were not observed during the catalytic reaction. These
results indicate that I is stable under the reaction conditions.
In the case of divanadium-substituted POMs, [γ-XW10-
O38V2(μ-OH)2]n¹ (X = Si (n = 4) or P (n = 3)), the reaction of
the {V2(μ-OH)2} core with H2O2 gives the {V2(μ-OH)(μ-OOH)}
species followed by successive dehydration to form an active
oxygen species with the μ-η2:η2-O2 species.5 Similarly, protons
likely play an important role in the formation of the {Nb2(μ-
OH)2} site, which would react with H2O2 to form the bridging
peroxo species II, e.g. μ-η2:η2-O2, μ-η1:η1-O2 species.11
which agreed with the patterns calculated for [TMA6HSiW10-
O38Nb2(O2)2]+ and [TMA7HSiW10O38Nb2(O2)2]2+, respectively
(Figure S2a). The CSI-MS spectra of TBA5I and TPeA5I
showed similar intense signal sets (Figures S2b and S2c),
suggesting that the solution states of I are the same regardless of
the kind of cations. The 29Si NMR spectrum showed one signal
at ¹84.9 ppm (Figure 2a),9 suggesting that TPeA5I is a single
species. The 183W NMR spectrum showed six signals at ¹87.0,
¹105.6, ¹107.6, ¹126.2, ¹136.4, and ¹147.0 ppm with the
respective intensity ratio of 2:1:2:2:1:2,9 suggesting Cs symme-
try in CD3CN (Figure 2b). The stoichiometric oxidation of
triphenylphosphine (1 mmol) with TPeA5I (20 ¯mol) gave
39 ¯mol of triphenylphosphine oxide, showing that I has 2 equiv
of active oxygen species for phosphine oxidation. All these CSI-
MS and NMR results show that the solid-state structure of I is
maintained in the solution state.
To confirm the catalytic activity of I, the TBA5I-catalyzed
epoxidation of cyclooctene (1) with H2O2 was carried out under
various conditions (Table 1). In each case, 1,2-epoxycyclo-
octane (2) was mainly obtained with the formation of small
amounts of the corresponding diol (3), hydroxy ketone (4), and
enone (5). While TBA5I itself showed low catalytic activity for
the epoxidation, the reaction rates dramatically increased as the
amounts of HClO4 increased up to 2 equiv and did not change
much upon addition of more than 2 equiv of HClO4. The yield
of 2 was also dependent on the amounts of HClO4, and the yield
of 2 reached up to 56% when using 2 equiv of HClO4. The
same dependence was also observed for the TPeA5I-catalyzed
epoxidation of 1 (Figure S4). In addition, the reaction rate for
the catalytic epoxidation by TBA5I with H2O2 in the presence of
2 equiv of HClO4 (5.4 © 10¹2 mM min¹1) was much larger than
those for (i) the catalytic epoxidation by TBA5I with H2O2 in the
absence of HClO4 (1.7 © 10¹3 mM min¹1) and (ii) the stoichio-
metric epoxidation with TBA5I in the presence of 2 equiv of
HClO4 (4.3 © 10¹3 mM min¹1) (Figure S5). These results show
that I requires both protons and H2O2 to achieve high catalytic
performance, and that the further reaction of I with H2O2 likely
proceeds in the presence of protons to give the active species for
epoxidation.
Finally, the scope of the present TBA5I-catalyzed oxidation
system was examined (Table 2). Cyclooctene selectively gave
1,2-epoxycyclooctane. The oxidation of thioanisole exclusively
900 | Chem. Lett. 2015, 44, 899–901 | doi:10.1246/cl.150213
© 2015 The Chemical Society of Japan

Table 2. Oxidation of various substrates with H2O2 by TBA5I in the
presence of HClO4
2
a) C. L. Hill, C. M. Prosser-McCartha, Coord. Chem. Rev.
1995, 143, 407. b) T. Okuhara, N. Mizuno, M. Misono, Adv.
Catal. 1996, 41, 113. c) N. Mizuno, K. Yamaguchi, K.
Kamata, Coord. Chem. Rev. 2005, 249, 1944. d) Modern
Heterogeneous Oxidation Catalysis: Design, Reactions
and Characterization, ed. by N. Mizuno, Wiley-VCH,
Weinheim, 2009. doi:10.1002/9783527627547. e) N.
Mizuno, K. Kamata, Coord. Chem. Rev. 2011, 255, 2358.
f) J.-Y. Piquemal, E. Briot, J.-M. Brégeault, Dalton Trans.
2013, 42, 29.
a) B. S. Lane, K. Burgess, Chem. Rev. 2003, 103, 2457.
b) Mechanisms in Homogeneous and Heterogeneous
Epoxidation Catalysis, ed. by S. T. Oyama, Elsevier,
Amsterdam, 2008.
a) Y. Goto, K. Kamata, K. Yamaguchi, K. Uehara, S.
Hikichi, N. Mizuno, Inorg. Chem. 2006, 45, 2347. b) C. N.
Kato, K. Hayashi, S. Negishi, K. Nomiya, J. Mol. Catal. A:
Chem. 2007, 262, 25.
a) Y. Nakagawa, K. Kamata, M. Kotani, K. Yamaguchi, N.
Mizuno, Angew. Chem., Int. Ed. 2005, 44, 5136. b) Y.
Nakagawa, N. Mizuno, Inorg. Chem. 2007, 46, 1727. c) K.
Kamata, K. Yonehara, Y. Nakagawa, K. Uehara, N. Mizuno,
Nat. Chem. 2010, 2, 478. d) K. Yonehara, K. Kamata, K.
Yamaguchi, N. Mizuno, Chem. Commun. 2011, 47, 1692.
e) K. Kamata, K. Sugahara, K. Yonehara, R. Ishimoto, N.
Mizuno, Chem.®Eur. J. 2011, 17, 7549. f) K. Kamata, T.
Yamaura, N. Mizuno, Angew. Chem., Int. Ed. 2012, 51,
7275.
a) M. W. Droege, R. G. Finke, J. Mol. Catal. 1991, 69, 323.
b) M. K. Harrup, G.-S. Kim, H. Zeng, R. P. Johnson, D.
VanDerveer, C. L. Hill, Inorg. Chem. 1998, 37, 5550.
D. A. Judd, Q. Chen, C. F. Campana, C. L. Hill, J. Am.
Chem. Soc. 1997, 119, 5461.
a
Time
/h
Product
Yield
/%
Entry
1
Substrate
(Selectivity/%)
24
60
97
O
(94)
O
S
S
2
0.17
(98)
O
3
4
5
OH
3
4
24
72
89
65
(99)
O
OH
OH
O
(50)
(23)
(27)
OH
HO
OH
aReaction conditions: TBA5I (4 ¯mol), substrate (1.0 mmol), 30%
H2O2 (0.1 mmol), HClO4 (2 equiv with respect to TBA5I), CH3CN
(3 mL), 283 K. Yield and selectivity were determined by GC or
1H NMR analysis. Yield (%) = total products (mol)/initial H2O2
(mol) © 100. Selectivity (%) = desired product (mol)/total products
(mol) © 100.
gave the corresponding sulfoxide in 95% yield. In this case, the
corresponding sulfone was hardly produced. 1-Phenylethanol
was also selectively oxidized to acetophenone in 89% yield.
Allyl alcohol was oxidized to the corresponding epoxide (50%
selectivity), aldehyde (23% selectivity), and triol (27% selectiv-
ity). This selectivity for the TBA5I-catalyzed oxidation of allyl
alcohol is very different from those for triniobium-substituted-
POM-catalyzed ones.6 To the best of our knowledge, efficient
H2O2-based oxidation by niobium-substituted POM catalysts
applicable to various organic substrates is unprecedented.
In conclusion, a novel diniobium-substituted silicodeca-
tungstate with two η2-coordinated peroxo groups, [γ-HSi-
W10O38Nb2(η2-O2)2]5¹ (I) was successfully synthesized and
characterized. The 29Si and 183W NMR, and CSI-MS spectra
suggest that I reacts with H2O2 in the presence of protons
to form the possible active species II. Compound I could
efficiently catalyze the H2O2-based oxidation of various organic
substrates in the presence of a suitable proton source.
6
7
8
a) D. Bayot, B. Tinant, B. Mathieu, J.-P. Declercq, M.
Devillers, Eur. J. Inorg. Chem. 2003, 737. b) D. Bayot, B.
Tinant, M. Devillers, Catal. Today 2003, 78, 439. c) S.-J. Li,
S.-X. Liu, C.-C. Li, F.-J. Ma, D.-D. Liang, W. Zhang, R.-K.
Tan, Y.-Y. Zhang, L. Xu, Chem.®Eur. J. 2010, 16, 13435.
d) Q. Geng, Q. Liu, P. Ma, J. Wang, J. Niu, Dalton Trans.
2014, 43, 9843.
9
The 29Si and 183W NMR spectra of I were measured by using
TPeA5I because the solubilities of TMA5I and TBA5I in
CD3CN were below the detection limits of 29Si and 183W
nuclei.
10 a) G. Wahl, D. Kleinhenz, A. Schorm, J. Sundermeyer, R.
Stowasser, C. Rummey, G. Bringmann, C. Fickert, W.
Kiefer, Chem.®Eur. J. 1999, 5, 3237. b) A. Hroch, G.
Gemmecker, W. R. Thiel, Eur. J. Inorg. Chem. 2000, 1107.
c) D. V. Deubel, J. Sundermeyer, G. Frenking, Org. Lett.
2001, 3, 329. d) R. Ishimoto, K. Kamata, N. Mizuno, Angew.
Chem., Int. Ed. 2012, 51, 4662.
11 Kinetic studies showed the first-order dependence of the
reaction rates on the concentration of H2O2 (8.33-49.8 mM,
Figure S7b). The dependence of the reaction rates on the
concentration of 1 (0-1.33 M) showed saturation kinetics
(Figure S7c) which can be explained by the change of the
rate-determining step from the oxygen transfer step from II
to the C=C double bond (at low concentration of 1) to the
regeneration of II with H2O2 (at high concentration of 1) in
a similar way to those of divanadium-substituted POMs.5
Supporting Information is available electronically on J-STAGE.
References and Notes
³
Present address: Materials and Structures Laboratory, Tokyo
Institute of Technology, 4259 Nagatsuta-cho, Midori-ku,
Yokohama, Kanagawa 226-8503
1
a) C. L. Hill, Chem. Rev. 1998, 98, 1. b) M. T. Pope, in
Comprehensive Coordination Chemistry II, ed. by J. A.
McCleverty, T. J. Meyer, Elsevier Pergamon, Amsterdam,
2003, Vol. 4, pp. 635-678. doi:10.1016/B0-08-043748-6/
03035-8. c) A. Proust, R. Thouvenot, P. Gouzerh, Chem.
Commun. 2008, 1837. d) D.-L. Long, R. Tsunashima, L.
Cronin, Angew. Chem., Int. Ed. 2010, 49, 1736. e) L. Cronin,
A. Müller, Chem. Soc. Rev. 2012, 41, 7333.
Chem. Lett. 2015, 44, 899–901 | doi:10.1246/cl.150213
© 2015 The Chemical Society of Japan | 901