Synthesis and oxidation catalysis of a Ti-substituted phosphotungstate, and identification of the active oxygen species

Article abstract of DOI:10.1039/c5cy01031d

In this paper, we report the synthesis of a Ti-substituted phosphotungstate, TBA₆[(γ-PW₁₀O₃₆)₂Ti₄(μ-O)₂(μ-OH)₄] (I, TBA = tetra-n-butylammonium), and its application to H₂O₂-based oxidation. Firstly, an organic solvent-soluble dilacunary phosphotungstate precursor, TBA₃[γ-PW₁₀O₃₄(H₂O)₂] (PW10), has been synthesized. By the reaction of PW10 and TiO(acac)₂ (acac = acetylacetonate) in an organic medium (acetonitrile), I can be obtained. Compound I possesses a tetranuclear Ti core which can effectively activate H₂O₂ and shows high catalytic performance for several oxidation reactions, such as epoxidation of alkenes, oxygenation of sulfides, oxidative bromination of unsaturated compounds, and hydroxylation of anisole, giving the corresponding oxidation products with high efficiencies and selectivities. The catalytic performance of I is much superior to those of previously reported Ti-substituted polyoxometalates. In addition, I is highly durable during catalysis and can be reused several times while keeping its high catalytic performance. Furthermore, we have successfully isolated the truly catalytically active species for the present I-catalyzed oxidation, TBA₆[(γ-PW₁₀O₃₆)₂Ti₄(μ-η²:η²-O₂)₄] (II), and its anion structure has been determined by X-ray crystallographic analysis. All of the four Ti₂-μ-η²:η²-peroxo species in II are active for stoichiometric oxidation (without H₂O₂), and II is included in the catalytic cycle for I-catalyzed oxidation.

Full text of DOI:10.1039/c5cy01031d

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Synthesis and oxidation catalysis of a Ti-
substituted phosphotungstate, and identification
Cite this: DOI: 10.1039/c5cy01031d
of the active oxygen species†
Eri Takahashi,a Keigo Kamata,‡a Yuji Kikukawa,§a Sota Sato,b Kosuke Suzuki,a
a
a
*
Kazuya Yamaguchi and Noritaka Mizuno
In this paper, we report the synthesis of a Ti-substituted phosphotungstate, TBA6ij(γ-PW10O36)2Ti4IJμ-O)2IJμ-
OH)4] (I, TBA = tetra-n-butylammonium), and its application to H2O2-based oxidation. Firstly, an organic
solvent-soluble dilacunary phosphotungstate precursor, TBA3ijγ-PW10O34IJH2O)2] (PW10), has been synthe-
sized. By the reaction of PW10 and TiOIJacac)2 (acac = acetylacetonate) in an organic medium (acetonitrile),
I can be obtained. Compound I possesses a tetranuclear Ti core which can effectively activate H2O2 and
shows high catalytic performance for several oxidation reactions, such as epoxidation of alkenes, oxygena-
tion of sulfides, oxidative bromination of unsaturated compounds, and hydroxylation of anisole, giving the
corresponding oxidation products with high efficiencies and selectivities. The catalytic performance of I is
much superior to those of previously reported Ti-substituted polyoxometalates. In addition, I is highly dura-
ble during catalysis and can be reused several times while keeping its high catalytic performance. Further-
more, we have successfully isolated the truly catalytically active species for the present I-catalyzed oxida-
tion, TBA6ij(γ-PW10O36)2Ti4IJμ-η2:η2-O2)4] (II), and its anion structure has been determined by X-ray
crystallographic analysis. All of the four Ti2-μ-η2:η2-peroxo species in II are active for stoichiometric oxida-
tion (without H2O2), and II is included in the catalytic cycle for I-catalyzed oxidation.
Received 8th July 2015,
Accepted 3rd August 2015
DOI: 10.1039/c5cy01031d
www.rsc.org/catalysis
waste in laboratory-scale as well as industrial-scale produc-
tion, antiquated stoichiometric oxidation processes should be
Introduction
Selective oxidation is one of the most prevalent transforma-
tions in a broad range of fields, and the oxygenated products,
such as epoxides, alcohols, aldehydes, ketones, and carboxylic
acids, greatly contribute to the quality of our lives. However,
oxidation processes sometimes lead to vast amounts of waste;
for example, dichromate, permanganate, and peracids are still
frequently utilized in (super)stoichiometric amounts espe-
cially for fine chemicals production. In order to minimize
replaced with catalytic ones.1 The choice of oxidant is also
very important, and H2O2 and O2 (or air) are regarded as “gre-
ener oxidants” because of their high contents of active oxygen
species, e.g., 47 wt% in the case of H2O2, and co-production
of only water.1 Therefore, the current most important subject
of basic research is to develop reliable catalysts that can acti-
vate the greener oxidants and transfer the active oxygen spe-
cies to various substrates with high efficiency and selectivity.
Up to the present, a number of transition-metal catalysts
have been developed for H2O2-based oxidation of various
substrates.2–10 The activation of H2O2 by transition-metal cata-
lysts typically proceeds through a one-electron (Fenton-like)
mechanism, a two-electron mechanism, or both.11 Unlike the
one-electron mechanism, free-radical intermediates are not
involved, competitive non-productive H2O2 decomposition is
suppressed, and electrophilically activated metal-IJhydro)-
peroxo and/or metal-oxo species are recognized to be formed
in the two-electron mechanism.11 Thus, the two-electron
mechanism is desirable for H2O2-based selective oxidation
especially for electrophilic oxygen transfer reactions to
organic substrates. Generally, Ti, V, Mo, W, and Re in their d0
electronic configurations are the principal metals that can
a Department of Applied Chemistry, School of Engineering, The University of
Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan.
E-mail: tmizuno@mail.ecc.u-tokyo.ac.jp
b JST, ERATO, Isobe Degenerate π-Integration Project, Japan; Advanced Institute
for Materials Research (AIMR) and Department of Chemistry, Tohoku University,
Aoba-ku, Sendai 980-8577, Japan
† Electronic supplementary information (ESI) available. CCDC 1406336–1406337.
For ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c5cy01031d
‡ Present address: Materials and Structures Laboratory, Tokyo Institute of
Technology, Nagatsuta-cho 4259, Midori-ku, Yokohama-city, Kanagawa 226-
8503, Japan.
§ Present address: Department of Chemistry, Division of Material Sciences,
Graduate School of Natural Science and Technology, Kanazawa University,
Kakuma-machi, Kanazawa 920-1192, Japan.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
catalyze H2O2-based oxidation through the two-electron
mechanism.11
“structural motifs”.14 To date, various kinds of metal-substituted
POMs have been synthesized for H2O2-based oxidation.14,15
With regard to the Ti-substituted ones, ij(γ-SiW10O36)2Ti4IJμ-
Amongst d0-metal catalysts for H2O2-based oxidation, Ti-
based catalysts have extensively been investigated.2–4 Since
the first report on the synthesis and oxidation catalysis of TS-
1 (titanium silicalite-1),2a numerous Ti-containing zeolites
have been synthesized by isomorphous substitution for Si (or
Al) in the framework.2 The tetrahedrally coordinated Ti site
in TS-1 is believed to contribute the oxidation catalysis, and
TiIJη1-OOH), TiIJη2-OOH), and/or TiIJη2-O2) species (with coordi-
nation of an alcohol (solvent) or a water molecule to the Ti
site) have been postulated as active oxygen species on the
basis of both experimental approaches and computational
calculations (Fig. 1a).2,12 Recently, Katsuki and co-workers
have reported the synthesis of bis-μ-oxo Ti(salen) and
Ti(salan) dimeric complexes and their utilization for enantio-
selective epoxidation with H2O2.3a,b Soon after, they have suc-
cessfully synthesized a μ-oxo-μ-η2:η2-peroxo Ti(salan) complex
by the reaction of the bis-μ-oxo complex with H2O2, and its
structure has been determined by X-ray crystallographic
analysis.3c Although the μ-oxo-μ-η2:η2-peroxo complex is effec-
tive for catalytic epoxidation with H2O2, it is completely inac-
tive for stoichiometric epoxidation. Thus, the complex is not
a truly catalytically active species but just a “precursor” of the
active species (Fig. 1b).3c As stated above, although Ti cata-
lysts are recognized to be effective for H2O2-based oxidation,
the truly catalytically active species as well as the detailed
reaction mechanism are still unclear and controversial.
We have engaged in the design of polyoxometalate (POM)-
based molecular catalysts for a long time.13 POMs are a large
class of metal–oxygen cluster anions, and their properties,
such as redox potentials, acidic and basic properties, and sol-
ubilities, can be finely tuned by choosing constituent ele-
ments and counter cations.14 In particular, POMs are attrac-
tive materials for oxidation catalysts because they are
thermally and oxidatively stable, and active site structures
can precisely be designed by using lacunary POMs as
O)2IJμ-OH)4]8−,4a ijPW11TiO40]5−,4b–d ijPW10Ti2O40]7−,4b ij(α-1,2,3-
7− 4f
PW9Ti3O37)2IJμ-O)3]12− 4e
,
ijα-PW10{TiIJO2)}2O38
]
,
and ij(B-α-
AsW9O33)2{TiIJOH)}2{WOIJH2O)}]8– (ref. 4g) are the typical
examples. Despite these advances, there is still plenty of
room for improvement in their catalytic activities, and in
addition their truly catalytically active species have never
been clarified and isolated. A POM with Ti-peroxo species,
ijα-1,2,3-P2W15O56{TiIJη2-O2)}3IJμ-OH)3]9−, has also been synthe-
sized but this peroxo species is too stable to oxidize
substrates.16
In this study, we have synthesized for the first time a Ti-
substituted phosphotungstate with a tetranuclear Ti core,
TBA6ij(γ-PW10O36)2Ti4IJμ-O)2IJμ-OH)4] (I, TBA = tetra-n-
butylammonium), by the reaction of an organic solvent-soluble
divacant lacunary phosphotungstate, TBA3ijγ-PW10
(PW10), with an appropriate Ti4+ source IJTiOIJacac)2, acac =
acetylacetonate) in acetonitrile (CH3CN). Compound
O34IJH2O)2]
I
could act as an efficient reusable catalyst for several
H2O2-based oxidation reactions, such as epoxidation of
alkenes, oxygenation of sulfides, oxidative bromination of
unsaturated compounds, and hydroxylation of anisole. In
addition, we have successfully isolated the truly catalytically
active species, TBA6ij(γ-PW10O36)2Ti4IJμ-η2:η2-O2)4] (II), and its
anion structure has been determined by X-ray crystallo-
graphic analysis (Fig. 1c).
Results and discussion
Synthesis and structural characterization of the titanium-
substituted phosphotungstate
Generally, the introduction of metal cations into the vacant
sites of lacunary POMs has been performed in aqueous
media, and various metal-substituted POMs with unique
structures and properties have been synthesized.13–15 How-
ever, the synthesis in aqueous media requires strict control
of the synthetic conditions, such as pH, temperature, and
concentration, and POMs frequently isomerize to unexpected
structures.17 In addition, alkali metal cations (counter cat-
ions) strongly coordinate to the vacant sites of lacunary
POMs, which frequently inhibit the introduction of metal cat-
ions into the vacant sites.18 We have recently utilized organic
solvent-soluble lacunary POMs, e.g., TBA4ijγ-SiW10O34IJH2O)2]
(SiW10),8a as precursors and introduced metal cations into
their lacunary pockets in organic media.19 In the synthesis in
organic media, the undesirable isomerization of lacunary pre-
cursors can be avoided through metal introduction. We have
recently found that the reactivity of metal-substituted POMs
with H2O2 can be increased by changing the central hetero-
atoms from Si4+ to P5+, leading to the significant improve-
ment of catalytic activities for electrophilic oxidation; for
example, the reaction rate of the epoxidation of allyl acetate
by a V-substituted phosphotungstate, ijγ-PW10O38V2IJμ-OH)2]3−,
was ca. 16 times larger than that by the silicotungstate
Fig. 1 (a) One of the postulated active oxygen species formed on TS-
1,2 (b) μ-oxo-μ-η2:η2-peroxo core in the Katsuki's Ti(salan) complex,3c
and (c) tetra-μ-η2:η2-peroxo core in II (this work).
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
analogue ijγ-SiW10O38V2IJμ-OH)2]4−.20 However, until now there
are only five examples (V, Mn, Mo, W, and Ru)21 of metal
substitution into ijγ-PW10O36]7−, likely owing to the low stabil-
ity and solubility of ijγ-PW10O36]7− (typically Cs salt) in aque-
ous media.
In order to demonstrate metal substitution in organic
media, organic solvent-soluble ijγ-PW10O36]7− precursors (e.g.,
TBA salt) are required, but they are unprecedented com-
pounds. Therefore, we first attempted to synthesize an
organic solvent-soluble divacant lacunary phosphotungstate
precursor (PW10) by the cation exchange reaction of Cs7ijγ-
PW10O36] (ref. 22) with TBABr. To avoid the undesirable isom-
erization of the POM framework, the reaction was carefully
carried out at low temperature (273 K). The IR spectrum of
PW10 showed bands around 1109–1036, 975–951, and 861–
798 cm−1 assignable to νIJP–O), νIJWO), and νIJW–O–W),
respectively, and these bands were also observed at similar
regions in the case of Cs7ijγ-PW10O36] (Fig. S1a and b†). The
31P NMR spectrum of PW10 in DMSO-d6 showed a single sig-
nal at −10.5 ppm (Fig. 2a). The 183W NMR spectrum of PW10
in DMSO-d6 showed five signals at −105.4, −114.2, −117.3,
−136.1, and −168.5 ppm with the respective intensity ratio of
1 : 1 : 1 : 1 : 1, indicating that PW10 possesses the C2 symmetry
in the solution-state which is the same as that of the
silicotungstate analogue SiW10 (Fig. 3a).8a The cold-spray
ionization mass (CSI-MS) spectrum of PW10 in acetone
showed the signal set at m/z = 3384 with isotopic distribution
which agreed with the pattern calculated for ijTBA4IJPW10O34)]+
(Fig. S2a†). The elemental analysis data showed a TBA/P/W
ratio of 3 : 1 : 10. Therefore, the formula of PW10 is possibly
TBA3ijγ-PW10O34IJH2O)2].
Fig. 3 183W NMR spectra of (a) PW10 in DMSO-d6 and (b) I in CD3CN/
PC (1 : 1 v/v).
determined. The structure of the anion part of I is shown in
Fig. 4, and the crystallographic data are summarized in Table
S1.† An edge-shared oxygen-bridged dinuclear titanium core
was introduced into the lacunary pocket of ijγ-PW10O36]7−, and
two monomer subunits were linked by two Ti–O–Ti frag-
ments. Six TBA cations per I anion could crystallographically
be assigned in accord with the elemental analysis data. The
bond valence sum (BVS) values of phosphorus (4.96), tung-
sten (5.98–6.15), and titanium (4.22–4.24) atoms in I indicate
that their respective valences are +5, +6, and +4. Therefore,
four protons are associated with the I anion. The BVS values
of O1a and O1b were 1.23 and 1.27, respectively, suggesting
that these oxygens were monoprotonated to form the ijTi2IJμ-
OH)2]6+ core in the subunit. The BVS value of O1c was 2.10,
showing that the bridging oxygen atom was a μ-O species.
The X-ray crystallographic and elemental analysis data show
that the formula of I is TBA6ij(γ-PW10O36)2Ti4IJμ-O)2IJμ-OH)4].
The formation of I can be expressed by the following equa-
tion [eqn (1)].
By using PW10 as
a
precursor,
a
Ti-substituted
phosphotungstate (I) could successfully be synthesized. Com-
pound I could readily be synthesized by just simply mixing
the required components of PW10 and a suitable Ti source
IJTiOIJacac)2, two equivalents with respect to PW10) in CH3CN
at room temperature. Fortunately, the single crystals of I suit-
able for X-ray crystallographic analysis were obtained by
recrystallization from a mixed solvent of CH3CN and ethyl
acetate (EtOAc), and its structure was successfully
2TBA3[γ ‐ PW10O34(H2O)2] + 4TiO(acac)2 + 2H2O
→ TBA6[(γ ‐ PW10O36)2Ti4(μ ‐ O)2(μ ‐ OH)4] + 8Hacac
(1)
Fig. 4 Polyhedral and ball-and-stick representation of the anion part
of I. The {WO6} and {PO4} units are shown as green octahedra and pur-
ple tetrahedra, respectively. Yellow and red spheres show Ti and O
atoms, respectively.
Fig. 2 31P NMR spectra of (a) PW10 in DMSO-d6 and (b) I in CD3CN.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
The anion structure of I was intrinsically isostructural with
mol% of I (with respect to H2O2) was used, 1,2-epoxycyclo-
octane was produced in 94% yield (Table 1, entry 5). The cata-
lytic activities of the precursors of I were much lower than
that of I; the epoxidation using PW10 and TiOIJacac)2 afforded
1,2-epoxycyclooctane in only 12% and 4% yields, respectively
(Table 1, entries 6 and 7). As mentioned above, I could read-
ily be formed by just mixing PW10 and TiOIJacac)2 in CH3CN.
Thus, it should be noted that a simple mixture of PW10 and
TiOIJacac)2 (the so-called “in situ-prepared catalyst”) could be
utilized for the epoxidation and that the in situ-prepared cata-
lyst showed almost the same performance as I (Table 1, entry
8). It would be very advantageous to demonstrate the reac-
tions using such in situ-prepared catalysts because there is
no need to synthesize (isolate) the catalysts separately.
Hence, several efficient POM-based in situ-prepared catalyst
systems in aqueous media25 as well as in organic media19
have been developed to date.
Up to the present, several Ti-substituted POMs for
H2O2-based oxidation have been reported.4 The catalytic per-
formance of I was much superior to those of previously
reported Ti-substituted POMs including our silicotungstate
analogue TBA8ij(γ-SiW10O36)2Ti4IJμ-O)2IJμ-OH)4] (Table S3†).4 As
we expected, the effect of the central heteroatoms was very
significant; the reaction rate of I under the conditions
described in Fig. S4† was 2.7 mM min−1 and ca. 11 times
larger than that of TBA8ij(γ-SiW10O36)2Ti4IJμ-O)2IJμ-OH)4] (0.26
mM min−1). By changing the central heteroatom from Si4+ to
P5+, the negative charge of the POM anion decreased, and
thus the catalytic activity for electrophilic oxidation likely
improved. This is one possible explanation for the effect of
the central heteroatoms, and such an effect is also observed
in the case of V-substituted POMs.20
those of the previously reported TBA8ij(γ-SiW10O36)2Ti4IJμ-O)2IJμ-
OH)4] (ref. 4a) and K8ij(γ-GeW10O36)2Ti4IJμ-O)2IJμ-OH)4].23 The
lengths of Ti⋯Ti (3.18–3.48 Å), Ti–OH (1.98–2.00 Å), and
Ti–O (1.79–1.80 Å), and the angles of Ti–OH–Ti (105.58–
106.58°) and Ti–O–Ti (151.09°) in I were quite similar to
those in TBA8ij(γ-SiW10O36)2Ti4IJμ-O)2IJμ-OH)4] and K8ij(γ-
GeW10O36)2Ti4IJμ-O)2IJμ-OH)4] (Table S2†). By contrast, the
lengths of Ti⋯P (3.77–3.79 Å) in I were slightly longer than
those of Ti⋯Si (3.61–3.69 Å) in TBA8ij(γ-SiW10O36)2Ti4IJμ-O)2IJμ-
OH)4] and Ti⋯Ge (3.60–3.61 Å) in K8ij(γ-GeW10O36)2Ti4IJμ-
O)2IJμ-OH)4].
The 31P NMR spectrum of I in CD3CN showed a single sig-
nal at −11.8 ppm, indicating the presence of I as the single
species (Fig. 2b). The 1H NMR spectrum of I in CD3CN
showed signals assignable to TBA cations (1.01, 1.43, 1.67,
and 3.17 ppm) and μ-OH species (7.91 ppm) (Fig. S3†). The
183W NMR spectrum of I in a mixed solvent of CD3CN and
propylene carbonate (PC, 1 : 1 v/v) showed three signals at
−110.6, −112.7, and −115.8 ppm with the respective intensity
ratio of 1 : 2 : 2 in accord with the C2v symmetry of I in the
solid state (Fig. 3b).24 The CSI-MS spectrum of I in CH3CN
showed two sets of signals centered at m/z = 3561 and 3674
assignable
to
ijTBA8H4{(PW10O36)2Ti4O6}]2+
and
ijTBA9H{(PW10O36)2Ti4O5}]2+, respectively (Fig. S2b†). These
NMR and CSI-MS spectra agreed well with the crystal struc-
ture of I, indicating that the solid-state structure of I was
maintained in the solution-state.
Catalytic performance for H2O2-based oxidations
To begin with, the epoxidation of cyclooctene with 30% aque-
ous H2O2 was carried out as the test reaction (Table 1). Under
the present reaction conditions, the epoxidation did not pro-
ceed at all in the absence of catalysts (Table 1, entry 1). In
the presence of I, cyclooctene was smoothly oxidized at 305 K
to give 1,2-epoxycyclooctane in 97% yield based on H2O2
(Table 1, entry 2). When the reaction temperature was
lowered to 273 K, the epoxidation still significantly proceeded
(Table 1, entry 3). Even under the stoichiometric conditions
using one equivalent of H2O2 with respect to cyclooctene, 1,2-
epoxycyclooctane was obtained in 94% yield (Table 1, entry
4). The amount of I could be much reduced; even when 0.1
During the catalytic oxidation, no degradation of I into
smaller molecular weight peroxotungstate species, such as
ijHPO4{WOIJO2)2}2]2− and ijPO4{WOIJO2)2}4]3−, was observed.
When an excess amount of diethyl ether was added to the
reaction solution, I dissolved in the solution could spontane-
ously precipitate out, and thus the used I after the reaction
could easily be recovered by simple filtration (precipitation
method, ≥90% recovery). Moreover, the recovered I could be
reused at least five times while keeping its high catalytic per-
formance for the same reaction (Fig. 5). It should be noted
that the IR, CSI-MS, and 31P NMR spectra of the recovered I
a
Table 1 Effects of catalysts on epoxidation of cyclooctene with H2O2
Entry
Catalyst (μmol)
Temp. (K)
Time (h)
Yield (%)
1
2
3
4b
5
6
7
8
None
I (4)
I (4)
I (4)
323
305
273
305
305
305
305
305
3
1
5
3.5
3.5
1
1
1
<1
97
90
94
94
12
4
I (1)
PW10 (8)
TiOIJacac)2 (16)
PW10 (8) + TiOIJacac)2 (16)
94
a
b
Reaction conditions: cyclooctene (5 mmol), 30% H2O2 (1 mmol), CH3CN (6 mL). Cyclooctene (1 mmol). Yields were determined by GC with
an internal standard. Yield (%) = epoxide (mol)/initial H2O2 (mol) × 100.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
catalyst even after the fifth reuse experiment were almost the
same as those of the as-prepared fresh I, showing the high
durability of I (Fig. S1d, S2c, and S5†).
due to the steric effect of the active oxygen species generated
on POMs.20
The present I-catalyzed oxidation system was also applica-
ble to the oxygenation of sulfides (Table 3). The reaction was
completed within 10 min and afforded the corresponding
oxygenated products in ≥90% total yields based on H2O2.
The oxygenation of thioanisole and its derivatives with
electron-donating as well as electron-withdrawing substitu-
ents on the phenyl rings efficiently proceeded (Table 3,
entries 1–5). With regard to 4-cyanothioanisole, the reaction
exclusively took place at the sulfide group, and no Payne-type
reaction occurred at the cyano group (Table 3, entry 5). Not
only the aryl sulfides but also the less reactive alkyl ones
could effectively be oxygenated (Table 3, entries 6 and 7). Fur-
thermore, I could efficiently catalyze the oxidative bromina-
tion of various types of unsaturated compounds such as
arenes, alkenes, and alkynes to give the corresponding mono-
or dibrominated compounds in high yields, and the selected
examples are summarized in Scheme 1. The hydroxylation of
anisole could also be accomplished by using the present sys-
tem and preferentially took place at the para-position
(Scheme 2). As can be seen from the above-mentioned
results, one may recognize the broad synthetic utility of the
present I-catalyzed oxidation system.
Next, the synthetic scope of the present I-catalyzed oxida-
tion system was investigated. In the presence of catalytic
amounts of I, epoxidation of alkenes, oxygenation of sulfides,
oxidative bromination of unsaturated compounds, and
hydroxylation of anisole with H2O2 efficiently proceeded to
give the corresponding oxidation products. As for epoxida-
tion, various kinds of structurally diverse alkenes including
cyclic, internal, and terminal ones could be converted into
the corresponding epoxides with high selectivities (Table 2).
In contrast to porous titanosilicate-based catalysts, the epoxi-
dation of relatively bulky alkenes, e.g., norbornene, cyclo-
octene and cyclododecene, efficiently proceeded (Table 2,
entries 1–3). A non-activated terminal alkene of 1-octene
could also be converted into the corresponding epoxide
(Table 2, entry 8). For the epoxidation of cis- and
trans-alkenes, the configurations around the CC were
completely retained in the corresponding epoxides (Table 2,
entries 5–7), indicating no involvement of free-radical inter-
mediates in the present I-catalyzed epoxidation. In the epoxi-
dation of cis- and trans-2-octenes, the yield of cis-2,3-
epoxyoctane was significantly higher than that of trans-2,3-
epoxyoctane, while the πIJCC) HOMO energy of cis-2-octene
(−0.342 eV) calculated at the HF/6-311GIJd,p) level was almost
the same as that of trans-2-octene (−0.343 eV) (Table 2,
entries 5 and 6). Moreover, the epoxidation of 3-methyl-1-
cyclohexene was highly diastereoselective and afforded the
corresponding epoxide with the epoxy ring trans to the
methyl group (anti configuration) (Table 2, entry 4). Such
Identification of active oxygen species
With regard to our previously reported H2O2-based oxidation
catalyzed by TBA8ij(γ-SiW10O36)2Ti4IJμ-O)2IJμ-OH)4] (ref. 4a) and
TBAnijγ-XW10O38V2IJμ-OH)2] (X = Si (n = 4) or P (n = 3)),20 the
Ti2IJμ-OOH)IJμ-OH) and V2IJμ-η2:η2-O2) species have been postu-
lated as the respective active oxygen species on the basis of
the spectroscopic, kinetic, and/or computational evidence.
However, the isolation of the V2IJμ-η2:η2-O2) species and even
the detection of the Ti2IJμ-OOH)IJμ-OH) species have not yet
been successful because their concentrations are very low
and/or they are not stable enough for isolation and/or
detection.
stereospecific
cis-preferential
and
diastereoselective
anti-preferential epoxidations are not common and are likely
We initially measured the NMR (31P and 183W) and CSI-
MS spectra of the in situ-formed active oxygen species by the
reaction of I with H2O2. As mentioned above, the 31P NMR
spectrum of I showed a single signal at −11.8 ppm (Fig. 6a).
Upon the addition of an excess amount of H2O2 (128 equiva-
lents with respect to I) to the CD3CN solution of I, the 31P
NMR signal due to I completely disappeared, and one new
signal appeared at −16.7 ppm (Fig. 6b). Upon the addition of
cyclooctene (128 equivalents with respect to I) to this solu-
tion, the signal at −16.7 ppm completely disappeared, and
the signal at −11.8 ppm due to I appeared again with the con-
comitant formation of 1,2-epoxycyclooctane (Fig. 6c). There-
fore, the −16.7-ppm signal would be due to the active oxygen
species (II) for the epoxidation. The 183W NMR spectrum of I
with H2O2 (64 equivalents with respect to I) in a mixed sol-
vent of CD3CN and PC (1 : 1 v/v) showed six signals at −94.3,
−110.2, −112.9, −118.7, −123.2, and −127.2 ppm with the
respective intensity ratio of 2 : 1 : 2 : 1 : 2 : 2 (Fig. 6f).24 The
Fig. 5 Reuse experiments for the epoxidation of cyclooctene with
H2O2; the profiles for I (blue diamond), and the first (red circle), second
(green triangle), third (yellow square), forth (purple cross), and fifth
(blue asterisk) reuse experiments. Reaction conditions: I (4 μmol),
cyclooctene (5 mmol), 30% H2O2 (1 mmol), CH3CN (6 mL), 305 K.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 2 Epoxidation of various alkenes with H2O2 catalyzed by Ia
Entry
1
Substrate
Product
Time (h)
3
Yield (%)
75(only exo)
2
1
2
97
3b
90(cis/trans = 12/88)
4
5
1
5
80(cis/trans = 23/77)
82
6c
7
9
3
43
70
8c
8
54
a
b
c
Reaction conditions: I (4 μmol), substrate (5 mmol), 30% H2O2 (1 mmol), CH3CN (6 mL), 305 K. Substrate: cis/trans = 30/70. 323 K. Yields
were determined by GC with an internal standard. Yield (%) = epoxide (mol)/initial H2O2 (mol) × 100.
Table 3 Oxygenation of various sulfides with H2O2 catalyzed by Ia
Yield (%)
Total
yield
(%)
Time
(min)
Entry
1
Substrate
Sulfoxide/sulfone
62/18
5
98
2
3
10
10
87/6
88/6
>99
>99
4
5
5
65/14
86/5
93
97
10
6
7
5
5
49/25
55/18
>99
90
a
Reaction conditions: I (2 μmol), substrate (1 mmol), 30% H2O2 (0.2 mmol), CH3CN (1 mL), 305 K. Yields were determined by GC with an
internal standard. Yield (%) = sulfoxide or sulfone (mol)/initial H2O2 (mol) × 100. Total yield (%) = {sulfoxide (mol) + sulfone (mol) × 2}/initial
H2O2 (mol) × 100.
183W NMR result indicates that II would possess the C2v or
C2h symmetry.
The 31P NMR spectrum of the in situ-formed II was almost
unchanged and no signals of low molecular weight
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
peroxotungstate fragments, such as ijHPO4{WOIJO2)2}2]2− and
ijPO4{WOIJO2)2}4]3− (typically observed at 2.8 and 4.5 ppm in
CD3CN, respectively8c), were observed even after the solution
was kept for 24 h at room temperature. These results clearly
indicate that II is highly stable and that the decomposition of
the ijγ-PW10O36]7− framework in I does not occur at all even in
the presence of excess amounts of H2O2. Due to the stability
as well as the relatively higher concentration of II, we
assumed that II can be isolated as the single species.
Scheme 1 Oxidative bromination of unsaturated compounds with
H2O2 catalyzed by I. Reaction conditions: I (0.25 μmol), substrate (1
mmol), NaBr (2 mmol), 30% H2O2 (1 mmol), AcOH/CH3CN IJ2/1 mL),
305 K. [a] NaBr (1 mmol). Yields were determined by GC with an
internal standard. Yield (%) = product (mol)/initial H2O2 (mol) × 100.
As we expected, II could readily be isolated by the addition
of an excess amount of diethyl ether to the CH3CN solution
containing I and H2O2 at 273 K. Furthermore, the single crys-
tals of II for X-ray crystallographic analysis were successfully
obtained by recrystallization of the isolated II from a mixed
solvent of nitroethane (EtNO2) and EtOAc at 278 K. The 31P
NMR spectrum of the single crystals in CD3CN was the same
as that of the in situ-formed II (Fig. 6d). The CSI-MS spectrum
in CH3CN showed the main set of signals centered at m/z =
3705 assignable to ijTBA8{(PW10O36)2Ti4IJO2)4}·TBAOH]2+ (Fig.
S2d†). Because the single crystals of II were quite easily dam-
aged by X-ray irradiation, the diffraction data were carefully
collected by continuously changing the irradiation spots at
the synchrotron facility (see the Experimental section for
details). Consequently, we managed to determine the anion
structure of II with four peroxo species. Although highly dis-
ordered TBA cations could not be crystallographically
assigned, the elemental analysis revealed the presence of six
TBA cations per II anion. The anion structure of II is shown
in Fig. 7, and the crystallographic data are summarized in
Table S1.† Two Ti atoms in the lacunary pocket of each ijγ-
PW10O36]7− subunit were linked by one μ-η2:η2-O2 species,
and two monomer subunits were linked by two μ-η2:η2-O2
species, that is, the presence of four Ti2-μ-η2:η2-peroxo species
in II. The anion structure of II exhibited C2v symmetry in
agreement with the 183W NMR spectrum of the in situ-formed
II (Fig. 6f). The elemental analysis data showed a TBA/P/W/Ti
ratio of 3 : 1 : 10 : 2. The stoichiometric oxidation of
triphenylphosphine with II (10 μmol) gave ca. four
Scheme
2 Hydroxylation of anisole with H2O2 catalyzed by I.
Reaction conditions: I (1.25 μmol), anisole (5 mmol), 30% H2O2 (0.1
mmol), CH3CN (2 mL), 333 K, 120 min. Yields were determined by GC
with an internal standard. Yield (%) = product (mol)/initial H2O2 (mol) ×
100.
Fig. 6 31P NMR spectra of (a) I, (b) in situ-formed II (I + 128H2O2), (c)
in situ-formed II + 128 equivalents of cyclooctene, and (d) isolated
single crystals of II in CD3CN; 183W NMR spectra of (e) I and (f) in situ-
formed II (I + 64H2O2) in CD3CN/PC (1 : 1 v/v).
Fig. 7 Polyhedral and ball-and-stick representation of the anion part
of II. The {WO6} and {PO4} units are shown as green octahedra and
purple tetrahedra, respectively. Yellow and red spheres show Ti and O
atoms, respectively.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
equivalents (42 μmol) of triphenylphosphine oxide based on
II, showing the presence of four active oxygen species in II.
These data indicate that the formula of II is TBA6ij(γ-
PW10O36)2Ti4IJμ-η2:η2-O2)4], and II can be formed by the reac-
tion of I with four equivalents of H2O2 [eqn (2)].
organic solvent-soluble lacunary precursor in an organic
medium demonstrated herein will be applicable to the syn-
thesis of various types of metal-substituted POMs and will
open up a new avenue for POM chemistry and synthesis. In
the presence of I, several H2O2-based oxidation reactions,
such as epoxidation of alkenes, oxygenation of sulfides, oxi-
dative bromination of unsaturated compounds, and hydroxyl-
ation of anisole, smoothly proceeded to afford the corre-
sponding oxidation products with high efficiencies and
selectivities, and the catalytic performance of I was much
superior to those of previously reported Ti-substituted
POMs including our silicotungstate analogue TBA8ij(γ-
SiW10O36)2Ti4IJμ-O)2IJμ-OH)4]. Moreover, I was highly durable
during catalysis (no degradation into smaller molecular
weight peroxotungstate species) and could be reused sev-
eral times while keeping its high catalytic performance.
The in situ-prepared catalyst synthesized by just mixing
PW10 and TiOIJacac)2 was also equally effective to I, which
would be advantageous for practical applications. Fortu-
nately, we could isolate the truly catalytically active species
for the present I-catalyzed H2O2-based oxidation (II), and its
structure could successfully be determined by X-ray crystal-
lographic analysis. Compound II possessed four Ti2-μ-η2:η2-
peroxo species, and all of the four peroxo species were
active for oxidation. To the best of our knowledge, II is the
first example of a structurally determined M2-μ-η2:η2-peroxo
species truly active for oxidation, which is included in the
catalytic cycle for the present I-catalyzed oxidation. We
hope that these findings will lead to new insight into the
chemistry of active oxygen species.
TBA6[(γ‐PW10O36)2Ti4(μ ‐ O)2(μ ‐ OH)4] + 4H2O2
⇄ TBA6[(γ ‐ PW10O36)2Ti4(μ ‐ η2 : η2 ‐ O2)4] + 6H2O
(2)
With regard to the μ-oxo-μ-η2:η2-peroxo Ti(salan) complex
reported by Katsuki and co-workers,3c one μ-η2:η2-O2 species,
and two N atoms and two O atoms in the salan ligand were
coordinated to each Ti atom in a pseudooctahedral arrange-
ment (Fig. 1b), resulting in the coordination number of each
Ti atom of seven. On the other hand, two O atoms in the
POM framework and two μ-η2:η2-O2 species were coordinated
to each Ti atom in a pseudotetrahedral arrangement and
there was no bridging μ-O ligand between Ti atoms in the
case of II. Thus, the coordination number of each Ti atom in
II was six. This is the most significant difference between II
and the inactive Katsuki's complex, and thus we assume that
the high catalytic performance of I would be ascribed to the
formation of the active pseudotetrahedral Ti peroxo species,
in good agreement with the high reactivity of tetrahedral tita-
nium species (within the zeolite framework).2,12
Besides the above phosphine oxidation, the stoichiometric
epoxidation of cyclooctene and thioanisole with II (4 μmol)
also gave ca. four equivalents of 1,2-epoxycyclooctane (17
μmol) and the corresponding oxygenated products (methy
phenyl sulfoxide: 7 μmol, methy phenyl sulfone: 5 μmol),
respectively, with respect to II, showing that all of the four
Ti2-μ-η2:η2-peroxo species in II were active for these oxida-
tions. Although to date several compounds with M2-μ-η2:η2- Experimental section
peroxo species (M = Ti,3c Zr,26a–c Hf,26a–c U,26d,e etc.) have
Materials
been synthesized and their structures have been determined,
TBABr, TiOIJacac)2, and EtNO2 were obtained from TCI, and
these peroxo species are not active for epoxidation. To the
best of our knowledge, II is the first example of a structurally
determined M2-μ-η2:η2-peroxo species which is truly active for
epoxidation. The reaction rate (0.043 mM min−1) of the “stoi-
chiometric” epoxidation of cyclooctene (0.035 M) with II (0.57
mM) agreed well with that (0.041 mM min−1) of the “cata-
lytic” epoxidation of cyclooctene (0.035 M) with H2O2 (0.14
M) by I (0.57 mM) (Fig. S6†), indicating that II is the truly
active oxygen species and that it is included in the catalytic
cycle for the present I-catalyzed oxidation.
used as received. HNO3, diethyl ether, CH3CN, EtOAc, 30%
H2O2, acetone, PC, methanol, and deuterated solvents
(CD3CN and DMSO-d6) were obtained from KANTO Chemical,
and used as received. Substrates were obtained from Sigma-
Aldrich, TCI, KANTO Chemical, and Wako Pure Chemical
Industries, and purified prior to use (if necessary).27 Cs7ijγ-
PW10O36] (ref. 22) and TBA8ij(γ-SiW10O36)2Ti4IJμ-O)2IJμ-OH)4]
(ref. 4a) were synthesized according to reported procedures.
Aqueous 90% H2O2 was prepared by concentration of 30%
H2O2.
Instruments
Conclusion
The IR spectra were measured on
a Jasco FT/IR-4100
In this work, we have obtained several important findings for
the POM-based catalyst design and the application to
H2O2-based oxidation. Firstly, we synthesized an organic
solvent-soluble divacant lacunary phosphotungstate (PW10)
for the first time. By the reaction of PW10 and TiOIJacac)2 in
CH3CN, a novel Ti-substituted phosphotungstate (I) could
readily be synthesized. The synthetic strategy using an
using KBr or KCl disks. The NMR spectra were measured at
room temperature on a JEOL JNM-EX270 (31P, 109 MHz) or a
JEOL JNM-ECA500 (1H, 500 MHz; 31P, 202 MHz; 183W, 20.6
MHz). The chemical shifts (δ) of 1H, 31P and 183W were
reported in ppm downfield from internal TMS, external 85%
H3PO4 (in D2O), and external 1 M Na2WO4 (in D2O), respec-
tively. NMR measurements were done in different solvents.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
The choice of solvent was simply based on solubility. The
cold-spray ionization mass (CSI-MS) spectra were recorded on
a JEOL JMS-T100CS. Typical measurements were as follows:
orifice voltage 135 V for positive ions; concentration 0.1 mM;
spray temperature 263 K; ion source at room temperature.
ICP-AES analyses were performed on a Shimadzu ICPS-8100.
Elemental analyses for C, H, and N were performed on a
Yanaco MT-6. GC analyses were performed on a Shimadzu
GC-2014 with a flame ionization detector (FID) equipped with
a Stabilwax capillary column (internal diameter = 0.25 mm,
length = 30 m). The GC-MS spectra were measured on a
Shimadzu GCMS-QP2010 at an ionization voltage of 70 eV.
HPLC analyses were performed on a Shimadzu Prominence
system with a UV-vis detector (Shimadzu SPD-20A, 270 nm)
equipped with a GL Science Inertsil ODS-3 column (particle
size = 5 μm, internal diameter = 4.6 mm ϕ, length = 250
mm), using a mixed solvent of methanol and water as the
eluent (9 : 1 v/v). The πIJCC) HOMO energies of cis- and
trans-2-octenes were calculated at the HF/6-311GIJd,p) level
with the Gaussian 09 program package.28
where B is a constant equal to 0.37 Å, and r′0 is the bond
valence parameter for a given atom pair.36
Catalytic oxidation
The catalytic oxidations were carried out in a 15 mL glass
tube reactor containing a magnetic stir bar. A typical proce-
dure was as follows: I, substrate, 30% H2O2, and CH3CN were
successively placed into the glass tube reactor. The reaction
mixture was stirred at 305 K. The yields of the products were
determined by GC analysis using an internal standard. All
products are known and confirmed by comparing their GC
retention times and/or GC-MS patterns with the authentic
samples.
Stoichiometric oxidation
The stoichiometric oxidations were carried out in a 15 mL
glass tube reactor containing a magnetic stir bar. A typical
procedure for the stoichiometric epoxidation of cyclooctene
was as follows: cyclooctene (0.25 mmol) and CH3CN (6.96
mL) were successively placed into the glass tube reactor. After
the solution was stirred at 253 K, the reaction was initiated
by addition of II (4 μmol). The reaction was periodically mon-
itored by GC. The reaction rates (R0) were determined from
the slopes of the reaction profiles ([epoxide] versus time) at
low conversions of cyclooctene (initial rate method). Finally,
ca. four equivalents of 1,2-epoxycyclooctane (17 μmol) were
produced. The stoichiometric oxidation of thioanisole was
carried out via the same procedure at room temperature, and
gave ca. four equivalents of oxygenated products (methy phe-
nyl sulfoxide: 7 μmol, methy phenyl sulfone: 5 μmol). A typi-
cal procedure for the stoichiometric oxidation of
triphenylphosphine was as follows: triphenylphosphine (0.25
mmol) and CH3CN (4 mL) were successively placed into a
glass tube reactor. After the solution was stirred at 305 K, the
reaction was initiated by addition of II (10 μmol). The reac-
tion was periodically monitored by HPLC. Finally, ca. four
equivalents (42 μmol) of triphenylphosphine oxide were
produced.
X-ray crystallography
Diffraction measurements of I were made on a Rigaku
MicroMax-007 Saturn 724 CCD detector with graphite mono-
chromated Mo Kα radiation (λ = 0.71069 Å) at 123 K. The dif-
fraction data of I were collected using CrystalClear.29 Diffrac-
tion measurements of II were made on the BL1A beamline in
KEK Photon Factory (synchrotron facility) with a diffractome-
ter equipped with a Dectris Pilatus 2M-F PAD detector (λ =
0.95000 Å) at 95 K. The diffraction data of II were collected
using a UGUI program. All the data were processed using
HKL2000.30 Neutral scattering factors were obtained from the
standard source. In the reduction of data, Lorentz and polari-
zation corrections were made. All structures were solved by
SHELXS-97 (direct methods) and refined by SHELXH-97.31
The
structural
analysis
was
performed
using
CrystalStructure,32 Win-GX,33 and Yadokari-XG.34 The metal
atoms (P, W, and Ti) in I were refined anisotropically. The
metal atoms (P, W, and Ti) and oxygen atoms in II were
refined anisotropically with SIMU. Some of the bond lengths
of II were refined with SADI and DFIX. The highly disordered
TBA cations of II could not be crystallographically assigned
and were omitted by using the SQUEEZE program of
PLATON,35 affording much improved analytical results.
CCDC-1406336 (I) and CCDC-1406337 (II) contain the supple-
mentary crystallographic data for this manuscript.
Synthesis of PW10
TBABr (0.67 g, 2.1 mmol) was dissolved in HNO3 (2.6 M, 200
mL) at 273 K. Then, Cs7ijγ-PW10O36]·H2O (2.0 g, 0.59 mmol)
was added in a single step, and the resulting solution was
stirred for 10 min. The resulting white precipitate of PW10
was collected by filtration and washed with water (100 mL)
and diethyl ether (200 mL) to afford the analytically pure
white powder of PW10 (1.6 g, 84% yield based on Cs7ijγ-
PW10O36], >99% purity). Compound PW10 was stored in a
freezer (at 253 K). To prevent the isomerization and/or
decomposition of PW10, vacuum drying and storage at room
temperature should be avoided. Elemental analysis calcd (%)
for C48H118N3O39PW10 IJTBA3ijPW10O34IJH2O)2]·3H2O): C, 17.84;
H, 3.68; N, 1.30; P, 0.96; W, 56.90. Found: C, 17.65; H, 3.53;
N, 1.21; P, 0.92; W, 56.50. IR (KBr pellet; 4000–400 cm−1):
3429, 2962, 2935, 2874, 1625, 1483, 1381, 1347, 1151, 1109,
BVS calculations
The BVS values were calculated by the expression for the vari-
ation of the length rij of a bond between two atoms i and j in
the observed crystal with valence Vi:
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
1057, 1036, 975, 951, 861, 798, 675, 661, 613, 535, 401 cm−1.
31P NMR (DMSO-d6): δ –10.5 ppm. 183W NMR (DMSO-d6): δ –
105.4, −114.2, −117.3, −136.1, −168.5 ppm (the integrated
intensity ratio of 1 : 1 : 1 : 1 : 1). CSI-MS (acetone): m/z = 3384
ijTBA4IJPW10O34)]+.
759, 524, 459, 424 cm−1. 31P NMR (CD3CN): δ –16.9 ppm.
183W NMR (in situ, CD3CN/PC (1 : 1 v/v)): δ –94.3, −110.2,
−112.9, −118.7, −123.2, −127.2 ppm (the integrated intensity
ratio of 2 : 1 : 2 : 1 : 2 : 2). CSI-MS (CH3CN): m/z = 3705 (main
set of signals) ijTBA8{(PW10O36)2Ti4IJO2)4}·TBAOH]2+.
Synthesis of compound I
Acknowledgements
TiOIJacac)2 (0.16 g, 0.62 mmol) was dissolved in CH3CN (20
mL) at room temperature. Then, PW10 (1.0 g, 0.31 mmol)
was added in a single step, and the resulting solution was
stirred for 5 min. After filtration, the solution was added to
diethyl ether (200 mL) and the resulting pale yellow precipi-
tate was collected by filtration. The crude compound was
purified by recrystallization from a mixed solvent of CH3CN
and EtOAc. The solution was kept at room temperature for
24 h to afford pale yellow crystals of I (0.47 g, 45% yield
based on PW10). Elemental analysis calcd (%) for
We thank KEK Photon Factory (Research 2013G640) for the
use of the X-ray diffraction instruments (BL1A beamline). This
work was supported in part by the Japan Society for the Pro-
motion of Science (JSPS) through its “Funding Program for
World-Leading Innovative R&D on Science and Technology
(FIRST Program)” and Grants-in-Aid for JSPS Fellows and Sci-
entific Research from the Ministry of Education, Culture, Sci-
ence, Sports, and Technology of Japan.
C
100H230N6O81P2Ti4W20 IJTBA6ij(PW10O36)2Ti4O2IJOH)4]·EtOAc
Notes and references
·H2O): C, 17.81; H, 3.44; N, 1.25; P, 0.92; Ti, 2.84; W, 54.53.
Found: C, 18.25; H, 3.51; N, 1.15; P, 0.98; Ti, 2.53; W, 55.60.
IR (KBr pellet; 4000–400 cm−1): 3580, 3446, 2962, 2934, 2874,
1732, 1623, 1483, 1379, 1246, 1152, 1070, 1050, 1040, 971,
1 R. A. Sheldon, I. Arends and U. Hanefeld, Green Chemistry
and Catalysis, Wiley-VCH, Weinheim, 2007.
2 Ti (zeolites): (a) M. Taramasso, G. Perego and B. Notari, US
Pat. 4,410,501, 1983; (b) M. G. Clerici, Appl. Catal., 1991, 68,
249–261; (c) B. Notari, Catal. Today, 1993, 18, 163–172; (d) T.
Tatsumi, Metal-Substituted Zeolites as Heterogeneous
Oxidation Catalysts, Modern Heterogeneous Oxidation
Catalysis, ed. N. Mizuno, Wiley-VCH, Weinheim, 2009, pp.
125–155; (e) M. A. Camblor, A. Corma, A. Martínez and J.
956, 878, 815, 656, 537, 489, 420 cm−1 1H NMR (CD3CN): δ
.
7.91 (μ-OH, 4H), 4.06 (EtOAc, 2H), 3.17 (TBA, 48H), 1.97
(EtOAc, 3H), 1.67 (TBA, 48H), 1.43 (TBA, 48H), 1.20 (EtOAc,
3H), 1.01 (TBA, 72H) ppm. 31P NMR (CD3CN): δ –11.8 ppm.
183W NMR (CD3CN/PC (1 : 1 v/v)): δ –110.6, −112.7, −115.8
ppm (the integrated intensity ratio of 1 : 2 : 2). CSI-MS
(CH3CN): m/z
=
3561 ijTBA8H4{(PW10O36)2Ti4O6}]2+, 3674
Pérez-Pariente,
J.
Chem.
Soc.,
Chem.
Commun.,
ijTBA9H{(PW10O36)2Ti4O5}]2+, 6880 ijTBA7H4{(PW10O36)2Ti4O6}]+.
1992, 589–590; ( f ) P. Wu and Y. Tatsumi, Chem. Commun.,
2001, 897–898; (g) P. Wu, Y. Kubota and T. Yokoi, ACS
Catal., 2014, 4, 23–30.
Synthesis of compound II
3 Ti (salen and salan complexes): (a) K. Matsumoto, Y.
Sawada, B. Saito, K. Sakai and T. Katsuki, Angew. Chem., Int.
Ed., 2005, 44, 4935–4939; (b) Y. Sawada, K. Matsumoto, S.
Kondo, H. Watanabe, T. Ozawa, K. Suzuki, B. Saito and T.
Katsuki, Angew. Chem., Int. Ed., 2006, 45, 3478–3480; (c) S.
Kondo, K. Saruhashi, K. Seki, K. Matsubara, K. Miyaji, T.
Kubo, K. Matsumoto and T. Katsuki, Angew. Chem., Int. Ed.,
2008, 47, 10195–10198.
4 Ti (POMs): (a) Y. Goto, K. Kamata, K. Yamaguchi, K. Uehara,
S. Hikichi and N. Mizuno, Inorg. Chem., 2006, 45,
2347–2356; (b) T. Yamase, E. Ishikawa, Y. Asai and S. Kanai,
J. Mol. Catal. A: Chem., 1996, 114, 237–245; (c) L. Hua, Y.
Qiao, Y. Yu, W. Zhu, T. Cao, Y. Shi, H. Li, B. Feng and Z.
Hou, New J. Chem., 2011, 35, 1836–1841; (d) P. Jiménez-
Lozano, I. D. Ivanchikova, O. A. Kholdeeva, J. M. Poblet and
J. J. Carbó, Chem. Commun., 2012, 48, 9266–9268; (e) C. N.
Kato, K. Hayashi, S. Negishi and K. Nomiya, J. Mol. Catal. A:
Chem., 2007, 262, 25–29; ( f ) E. Ishikawa and T. Yamase,
J. Mol. Catal. A: Chem., 1999, 142, 61–76; (g) B. G. Donoeva,
T. A. Trubitsina, N. S. Antonova, J. J. Carbó, J. M. Poblet, G.
Al-Kadamany, U. Kortz and O. A. Kholdeeva, Eur. J. Inorg.
Chem., 2010, 5312–5317; (h) O. A. Kholdeeva, Eur. J. Inorg.
Chem., 2013, 1595–1605.
Compound I (0.10 g, 0.015 mmol) was dissolved in CH3CN (5
mL) at 273 K. Then, 90% H2O2 (128 equivalents, 1.9 mmol)
was added in a single step, and the resulting solution was
stirred for 1 h. After filtration, the solution was added to
diethyl ether (50 mL) and the resulting yellow precipitate was
collected by filtration and washed with EtOAc (20 mL). The
pure compound II was obtained as a yellow powder after dry-
ing (0.072 g, 72% yield based on I). Elemental analysis calcd
(%) for C100H228N6O84P2Ti4W20 IJTBA6ij(PW10O36)2Ti4IJO2)4]
·EtOAc·2H2O): C, 17.69; H, 3.38; N, 1.24; P, 0.91; Ti, 2.82; W,
54.16. Found: C, 17.10; H, 3.39; N, 1.20; P, 0.89; Ti, 2.81; W,
54.41. IR (KCl pellet; 4000–400 cm−1): 3410, 2960, 2932, 2872,
1624, 1482, 1466, 1379, 1131, 1062, 1041, 972, 955, 879, 816,
759, 524, 460, 422 cm−1. 31P NMR (CD3CN): δ –16.8 ppm. Pale
yellow crystals of II suitable for X-ray crystallographic analysis
were obtained by recrystallization from a mixed solvent of
EtNO2 and EtOAc at 278 K. Elemental analysis calcd (%) for
C102H233N7O86P2Ti4W20 IJTBA6ij(PW10O36)2Ti4IJO2)4]·EtNO2·EtOAc
·2H2O): C, 17.85; H, 3.42; N, 1.43; P, 0.90; Ti, 2.79; W, 53.57.
Found: C, 17.96; H, 3.44; N, 1.42; P, 0.90; Ti, 2.66; W, 53.36.
IR (KCl pellet; 4000–400 cm−1): 3421, 2961, 2933, 2873, 1732,
1624, 1546, 1483, 1379, 1131, 1062, 1040, 971, 955, 880, 818,
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
5 V: (a) A. Butler, M. J. Clague and G. E. Meister, Chem. Rev.,
1994, 94, 625–638; (b) V. Conte and B. Floris, Dalton Trans.,
2011, 40, 1419–1436; (c) D. Wischang, O. Brücher and J.
Hartung, Coord. Chem. Rev., 2011, 255, 2204–2217.
Xiong, J. Su, P. Li and H. Guo, J. Phys. Chem. C, 2012, 116,
9122–9131.
13 N. Mizuno, K. Yamaguchi and K. Kamata, Catal. Surv. Asia,
2011, 15, 68–79.
6 Mn :(a) P. Battioni, J. P. Renaud, J. F. Bartoli, M. Reina-
Artiles, M. Fort and D. Mansuy, J. Am. Chem. Soc., 1988, 110,
8462–8470; (b) D. E. De Vos, B. F. Sels, M. Reynaers, Y. V. S.
Rao and P. A. Jacobs, Tetrahedron Lett., 1998, 39, 3221–3224;
(c) B. S. Lane, M. Vogt, V. J. DeRose and K. Burgess, J. Am.
Chem. Soc., 2002, 124, 11946–11954; (d) P. Saisaha, J. W. de
Boer and W. R. Browne, Chem. Soc. Rev., 2013, 42,
2059–2074.
14 (a) C. L. Hill and M. Prosser-McCartha, Coord. Chem. Rev.,
1995, 143, 407–455; (b) N. Mizuno, K. Yamaguchi and K.
Kamata, Coord. Chem. Rev., 2005, 249, 1944–1956; (c) N.
Mizuno, K. Kamata, S. Uchida and K. Yamaguchi, Liquid-
Phase Oxidations with Hydrogen Peroxide and Molecular
Oxygen Catalyzed by Polyoxometalate-Based Compounds, Mod-
ern Heterogeneous Oxidation Catalysis, ed. N. Mizuno, Wiley-
VCH, Weinheim, 2009, pp. 185–216.
7 Fe: (a) M. Costas, K. Chen and L. Que, Jr., Coord. Chem.
Rev., 2000, 200–202, 517–544; (b) M. C. White, A. G. Doyle
and E. N. Jacobsen, J. Am. Chem. Soc., 2001, 123, 7194–7195;
(c) M. S. Chen and M. C. White, Science, 2007, 318, 783–787;
(d) E. P. Talsi and K. P. Bryliakov, Coord. Chem. Rev.,
2012, 256, 1418–1434.
15 (a) K. Kamata, K. Yonehara, Y. Nakagawa, K. Uehara and N.
Mizuno, Nat. Chem., 2010, 2, 478–483; (b) K. Kamata, T.
Yamaura and N. Mizuno, Angew. Chem., Int. Ed., 2012, 51,
7275–7278.
16 Y. Sakai, Y. Kitakoga, K. Hayashi, K. Yoza and K. Nomiya,
Eur. J. Inorg. Chem., 2004, 4646–4652.
8 W: (a) K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S.
Hikichi and N. Mizuno, Science, 2003, 300, 964–966; (b) R.
Noyori, M. Aoki and K. Sato, Chem. Commun.,
2003, 1977–1986; (c) K. Kamata, T. Hirano, S. Kuzuya and N.
Mizuno, J. Am. Chem. Soc., 2009, 131, 6997–7004; (d) R.
Ishimoto, K. Kamata and N. Mizuno, Angew. Chem., Int. Ed.,
2012, 51, 4662–4665; (e) C. Venturello, E. Alneri and M.
Ricci, J. Org. Chem., 1983, 48, 3831–3833; ( f ) Y. Ishii, K.
Yamawaki, T. Yoshida, T. Ura and M. Ogawa, J. Org. Chem.,
1987, 52, 1868–1870; (g) K. Sato, M. Aoki, M. Ogawa, T.
Hashimoto and R. Noyori, J. Org. Chem., 1996, 61,
8310–8311; (h) K. Sato, M. Aoki and R. Noyori, Science,
1998, 281, 1646–1647.
9 Re: (a) J. Rudolph, K. L. Reddy, J. P. Chiang and K. B.
Sharpless, J. Am. Chem. Soc., 1997, 119, 6189–6190; (b) W. A.
Herrmann, R. M. Kratzer, H. Ding, W. R. Thiel and H. Glas,
J. Organomet. Chem., 1998, 555, 293–295; (c) H. Adolfsson, C.
Copéret, J. P. Chiang and A. K. Yudin, J. Org. Chem.,
2000, 65, 8651–8658; (d) P. Altmann, M. Cokoja and F. E.
Kühn, Eur. J. Inorg. Chem., 2012, 3235–3239.
17 By the reaction of K8ijγ-SiW10O36] with TiOSO4 in acidic
solution (pH = 1.3) at room temperature, a dimeric Ti-
substituted γ-Keggin POM, ijTi4IJγ-SiW10O36)2IJμ-O)2IJμ-OH)4]8−
,
can be synthesized.4a On the other hand, using the same
starting materials, a tetrameric Ti-substituted POM com-
posed of β-Keggin subunits, ij(β-Ti2SiW10O39)4]24−, is obtained
even when the synthesis is performed at pH = 2 and 353 K.
See: F. Hussain, B. S. Bassil, L.-H. Bi, M. Reicke and U.
Kortz, Angew. Chem., Int. Ed., 2004, 43, 3485–3488.
18 H. Liu, J. Peng, Z. Su, Y. Chen, B. Dong, A. Tian, Z. Han and
E. Wang, Eur. J. Inorg. Chem., 2006, 4827–4833.
19 (a) Y. Kikukawa, K. Yamaguchi and N. Mizuno, Angew.
Chem., Int. Ed., 2010, 49, 6096–6100; (b) Y. Kikukawa, K.
Suzuki, M. Sugawa, T. Hirano, K. Kamata, K. Yamaguchi and
N. Mizuno, Angew. Chem., Int. Ed., 2012, 51, 3686–3690; (c)
Y. Kikukawa, Y. Kuroda, K. Yamaguchi and N. Mizuno,
Angew. Chem., Int. Ed., 2012, 51, 2434–2437; (d) T. Hirano, K.
Uehara, K. Kamata and N. Mizuno, J. Am. Chem. Soc.,
2012, 134, 6425–6433; (e) K. Suzuki, F. Tang, Y. Kikukawa, K.
Yamaguchi and N. Mizuno, Angew. Chem., Int. Ed., 2014, 53,
5356–5360; ( f ) K. Suzuki, M. Sugawa, Y. Kikukawa, K.
Kamata, K. Yamaguchi and N. Mizuno, Inorg. Chem.,
2012, 51, 6953–6961.
10 Pt: (a) M. Colladon, A. Scarso, P. Sgarbossa, R. A. Michelin
and G. Strukul, J. Am. Chem. Soc., 2006, 128, 14006–14007;
(b) M. Colladon, A. Scarso, P. Sgarbossa, R. A. Michelin and
G. Strukul, J. Am. Chem. Soc., 2007, 129, 7680–7689; (c) A. N.
Vedernikov, Acc. Chem. Res., 2012, 45, 803–813.
20 K. Kamata, K. Sugahara, K. Yonehara, R. Ishimoto and N.
Mizuno, Chem. – Eur. J., 2011, 17, 7549–7559.
11 (a) R. A. Sheldon and J. K. Kochi, Metal Catalyzed Oxidations
of Organic Compounds, Academic Press, New York, 1981; (b)
M. Hudlicky, Oxidations in Organic Chemistry, ACS
Monograph Series, American Chemical Society, Washington,
DC, 1990; (c) Modern Oxidation Methods, ed. J.-E. Bäckvall,
Wiley-VCH:, Weinheim, 2004.
12 (a) W. Y. Lin and H. Frei, J. Am. Chem. Soc., 2002, 124,
9292–9298; (b) S. Bordiga, A. Damin, F. Bonino, G.
Ricchiardi, C. Lamberti and A. Zecchina, Angew. Chem., Int.
Ed., 2002, 41, 4734–4737; (c) D. H. Wells, Jr., W. N. Delgass
and K. T. Thomson, J. Am. Chem. Soc., 2004, 126, 2956–2962;
(d) S. Bordiga, F. Bonino, A. Damin and C. Lamberti, Phys.
Chem. Chem. Phys., 2007, 9, 4854–4878; (e) L. Wang, G.
21 (a) P. J. Domaille and R. L. Harlow, J. Am. Chem. Soc.,
1986, 108, 2108–2109; (b) A. Patel and K. Patel, Polyhedron,
2014, 69, 110–118; (c) E. Cadot, V. Béreau and F. Sécheresse,
Inorg. Chim. Acta, 1996, 252, 101–106; (d) C. Besson, Z.
Huang, Y. V. Geletii, S. Lense, K. I. Hardcastle, D. G.
Musaev, T. Lian, A. Proust and C. L. Hill, Chem. Commun.,
2010, 46, 2784–2786.
22 P. J. Domaille, Inorg. Synth., 1990, 27, 96–104.
23 R. Tan, D. Li, H. Wu, C. Zhang and X. Wang, Inorg. Chem.
Commun., 2008, 11, 835–836.
24 Because the solubility of I in CD3CN was below the detection
limit of 183W nuclei, PC was added to dissolve I in the
solvent. The 31P NMR spectrum of I after the 183W NMR
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
measurement showed
Catalysis Science & Technology
a
single signal at −11.8 ppm,
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M.
Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J.
Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J.
Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian,
Inc., Wallingford CT, 2010.
suggesting that the structure of I was maintained during the
183W NMR measurement.
25 (a) I. A. Weinstock, E. M. G. Barbuzzi, M. W. Wemple, J. J.
Cowan, R. S. Reiner, D. M. Sonnen, R. A. Heintz, J. S. Bond
and C. L. Hill, Nature, 2001, 414, 191–195; (b) D. Sloboda-
Rozner, P. L. Alster and R. Neumann, J. Am. Chem. Soc.,
2003, 125, 5280–5281.
26 (a) S. S. Mal, N. H. Nsouli, M. Carraro, A. Sartorel, G.
Scorrano, H. Oelrich, L. Walder, M. Bonchio and U. Kortz,
Inorg. Chem., 2010, 49, 7–9; (b) D. Li, H. Han, Y. Wang, X.
Wang, Y. Li and E. Wang, Eur. J. Inorg. Chem.,
2013, 1926–1934; (c) B. S. Bassil, S. S. Mal, M. H. Dickman,
U. Kortz, H. Oelrich and L. Walder, J. Am. Chem. Soc.,
2008, 130, 6696–6697; (d) S. S. Mal, M. H. Dickman and U.
Kortz, Chem. – Eur. J., 2008, 14, 9851–9855; (e) P. Miró, J.
Ling, J. Qiu, P. C. Burns, L. Gagliardi and C. J. Cramer,
Inorg. Chem., 2012, 51, 8784–8790.
27 Purification of Laboratory Chemicals 5th ed., ed. D. D. Perrin
and W. L. F. Armarego, Pergamon Press, Oxford, 2003.
28 Gaussian 09, Revision C.01, M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H.
Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov,
J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr.,
J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J.
29 (a) CrystalClear 1.4.0;, Rigaku and Rigaku/MSC, The
Woodlands, TX, 1999; (b) J. W. Pflugrath, Acta Crystallogr.,
Sect. D: Biol. Crystallogr., 1999, 55, 1718–1725.
30 Z. Otwinowski and W. Minor, Processing of X-ray Diffraction
Data Collected in Oscillation Mode, Methods in Enzymology,
Macromolecular Crystallography, Part A, Academic Press, ed.
C. W. Carter, Jr. and R. M. Sweet, New York, 1997, vol. 276,
pp. 307–326.
31 G. M. Sheldrick, SHELX97, Programs for Crystal Structure
Analysis, Release 97-2, University of Göttingen: Göttingen,
Germany, 1997.
32 CrystalStructure 4.0, Rigaku and Rigaku/MSC, The
Woodlands, TX, 2000.
33 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
34 C. Kabuto, S. Akine, T. Nemoto and E. Kwon, Nippon Kessho
Gakkaishi, 2009, 51, 218–224.
35 P. van der Sluis and A. L. Spek, Acta Crystallogr., Sect. A:
Found. Adv., 1990, 46, 194–201.
36 I. D. Brown and D. Altermatt, Acta Crystallogr., Sect. B:
Struct. Sci., Cryst. Eng. Mater., 1985, 41, 244–247.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015