Investigation of the reaction mechanism for the epoxidation of alkenes with hydrogen peroxide catalyzed by a protonated tetranuclear peroxotungstate with nmr spectroscopy, kinetics, and DFT calculations

Article abstract of DOI:10.1002/ejic.201201058

For the epoxidation of cyclooctene with hydrogen peroxide (H ₂O₂), the catalytic activity of a dinuclear peroxotungstate, [{WO(O₂)₂}₂(μ-O)] ^2-, is strongly dependent on additives, and HClO₄ is the most effective. The reaction of [{WO(O₂)₂} ₂(μ-O)]^2- with HClO₄ (0.5 equiv.) gives a protonated tetranuclear peroxotungstate, [H{W₂O₂(O ₂)₄(μ-O)}₂]^3- (I). The diastereoselectivity for epoxidation of 3-methyl-1-cyclohexene shows that steric constraints of the active site of I are comparable to those of di- and tetranuclear peroxotungstates with XO₄^n- ligands (X = Se^VI, As^V, P^V, S^VI, and Si ^IV). The lowest X_SO [X_SO = (nucleophilic oxidation)/(total oxidation)] value of 0.13 for the I-catalyzed oxidation of thianthrene 5-oxide among peroxotungstates reveals that I has the most electrophilic active oxygen species. Kinetic and spectroscopic results show that an inactive species (I′) is reversibly formed by the reaction of I with water. Reaction rates for the catalytic epoxidation show first-order dependences on concentrations of alkene and I and a zero-order dependence on concentrations of H₂O₂. All these results indicate that oxygen transfer from I to a C=C double bond in an alkene is the rate-determining step. Computational studies support the proposed reaction mechanism. Copyright

Full text of DOI:10.1002/ejic.201201058

FULL PAPER
DOI:10.1002/ejic.201201058
Investigation of the Reaction Mechanism for the
Epoxidation of Alkenes with Hydrogen Peroxide Catalyzed
by a Protonated Tetranuclear Peroxotungstate with NMR
Spectroscopy, Kinetics, and DFT Calculations
CLUSTER
ISSUE
[
a]
[a]
[a]
Ryo Ishimoto, Keigo Kamata, and Noritaka Mizuno*
Dedicated to Professor Michael T. Pope on the occasion of his 80th birthday
Keywords: Homogeneous catalysis / Reaction mechanisms / Epoxidation / Peroxotungstates / Hydrogen peroxide
For the epoxidation of cyclooctene with hydrogen peroxide
for the I-catalyzed oxidation of thianthrene 5-oxide among
peroxotungstates reveals that I has the most electrophilic
active oxygen species. Kinetic and spectroscopic results show
that an inactive species (IЈ) is reversibly formed by the reac-
tion of I with water. Reaction rates for the catalytic epoxid-
(
[
2 2
H O ), the catalytic activity of a dinuclear peroxotungstate,
2–
{WO(O
is the most effective. The reaction of [{WO(O
O)] with HClO (0.5 equiv.) gives a protonated tetranuclear
peroxotungstate, [H{W
2
)
2
}
2
(μ-O)] , is strongly dependent on additives, and
HClO
4
2 2 2
) }
(μ-
2
–
4
3
–
2
O
2
(O
2
)
4
(μ-O)}
2
]
(I). The diastereo- ation show first-order dependences on concentrations of alk-
selectivity for epoxidation of 3-methyl-1-cyclohexene shows
ene and I and a zero-order dependence on concentrations of
. All these results indicate that oxygen transfer from I
to a C=C double bond in an alkene is the rate-determining
step. Computational studies support the proposed reaction
mechanism.
that steric constraints of the active site of I are comparable
2 2
H O
n–
to those of di- and tetranuclear peroxotungstates with XO
4
VI
V
V
VI
IV
ligands (X = Se , As , P , S , and Si ). The lowest XSO
SO = (nucleophilic oxidation)/(total oxidation)] value of 0.13
[
X
ated[7] systems, the reaction mechanism, including the rela-
tionship between additives and active species, has scarcely
been investigated.
Introduction
Epoxides are valuable compounds as resins, paints, sur-
factants, and intermediates in various organic syntheses.
In contrast to synthetic methods with low atom efficiencies
that use stoichiometric reagents, catalytic epoxidation of
alkenes with hydrogen peroxide (H O ) has attracted much
attention in both industry and academia because H O has
a high content of active oxygen species and only water is
formed as a byproduct. Therefore, a large number of H O -
based epoxidation systems that use transition-metal cata-
lysts have been developed.[ Since additives facilitate cata-
lyst tuning, the effects of additives have extensively been
investigated for H O -based epoxidation by metal com-
plexes and soluble metal oxides. Whereas some additives
such as amines, carboxylates, and acids enhance catalytic
[
1]
Polyoxometalates (POMs) are early transition-metal oxy-
gen anion clusters that have been applied to various fields
such as structural chemistry, analytical chemistry, surface
science, medicine, electrochemistry, photochemistry, and ca-
2
2
[8]
2
2
talysis. POMs have advantages as oxidation catalysts as
follows: (i) The redox and acid-base properties can be con-
trolled by changing the chemical compositions, (ii) they are
more oxidatively and thermally stable than organometallic
complexes, and (iii) catalytically active sites can be con-
trolled at atomic and/or molecular levels. Therefore, cata-
lytic H O -based oxidation reactions by various kinds of
2
2
2]
2
2
2
2
[9]
[10]
POMs such as peroxometalates, lacunary POMs,
and
[11]
transition-metal-substituted POMs have been developed.
We have previously reported the formation of a protonated
tetranuclear peroxotungstate, TPA [H{W O (O ) (μ-O)} ]
[
3]
activities and/or selectivities to epoxides for ruthenium-,
[
4]
[5]
[6]
manganese-,
iron-,
titanium-,
and tungsten-medi-
3
2
2
2 4
2
[
12]
(
TPA ·I; TPA = tetra-n-propylammonium; Figure 1),
3
[
a] Department of Applied Chemistry, School of Engineering, The and its highly efficient epoxidation of alkenes with H O ,
2 2
University of Tokyo,
although the reaction mechanism is still unclear.
7
-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
Fax: +81-3-5841-7220
E-mail: tmizuno@mail.ecc.u-tokyo.ac.jp
In this paper, we investigate the reaction mechanism for
the epoxidation of alkenes with H O catalyzed by TPA ·I
2
2
3
Homepage: http://park.itc.u-tokyo.ac.jp/mizuno/
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201201058.
with 17O NMR spectroscopy, kinetics, and density func-
tional theory (DFT) calculations including the effects of ad-
Eur. J. Inorg. Chem. 2013, 1943–1950
1943
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
(
(
Table 1, entry 2). The catalytic activity of TPA [{WO-
2
O ) } (μ-O)] was dramatically enhanced 50- to 60-fold
2
2 2
upon addition of HClO and HNO (Table 1, entries 1, 2,
4
3
and 14). However, Group 8–12 elements such as Fe, Co, Ni,
Zn, and Cd were moderately effective (Table 1, entries 3–8),
and Group 1 and 2 elements were not effective additives
(Table 1, entries 9–13).
2 2 2
Figure 1. Formation of I by the reaction of [{WO(O ) }
(μ-O)]2–
with 0.5 equiv. of a proton. Gray, red, and light-blue balls represent
tungsten, oxygen, and hydrogen atoms, respectively.
Steric and Electronic Nature of the Active Oxygen Species
in I
3-Alkyl-substituted cyclohexenes provide a quantitative
measure of purely steric effects on epoxidation reactions,
since electronic interactions between the C=C double bonds
ditives and the electronic and steric nature of the active oxy-
gen species.
[13]
and allylic substituents are not possible.
Therefore, the
steric character of the active oxygen species was investigated
by using 3-methyl-1-cyclohexene. The trans/cis ratio (66:34)
for epoxidation of 3-methyl-1-cyclohexene catalyzed by
Results and Discussion
TPA ·I [Equation (1)] was close to those catalyzed by
3
peroxotungstates such as [(n-C H ) N] [PO {WO(O ) } ]
6
13 4
3
4
2 2 4
Effects of Additives
[
[
10c]
10c]
(
(
56:44),
55:45),
[n-C H N(CH ) ] [{WO(O ) } (μ-O)]
12 25 3 3 2 2 2 2
and TBA [SeO {WO(O ) } ] (64:36, TBA =
Epoxidation of cyclooctene with H O catalyzed by
2
4
2 2 2
2
2
[
9j]
tetra-n-butylammonium), and lower than those by Ti-β
TPA [{WO(O ) } (μ-O)] in the presence of various kinds of
2
2 2 2
[
13a]
(
(
92:8)
and POMs such as TBA [γ-SiW O (H O) ]
additives was carried out (Table 1). The catalytic activity of
4 10 34 2 2
[
10c]
[11g]
81:19)
and TBA [γ-PV W O (μ-OH) ] (94:6),
3 2 10 38 2
TPA [{WO(O ) } (μ-O)] was strongly dependent on addi-
2
2 2 2
thus indicating that the steric effect of TPA ·I is comparable
tives. Among the additives tested, HClO was the most ef-
fective (Table 1, entry 1). We have reported that the reaction
3
4
to those of other peroxotungstates.
of TPA [{WO(O ) } (μ-O)] with HClO (0.5 equiv.) leads to
2
2 2 2
4
in situ formation of I and that the crystal structure of I is
a protonated tetranuclear peroxotungstate.[ Epoxidation
also efficiently proceeded with HNO3 instead of HClO4
12]
Table 1. Effects of additives on epoxidation of cyclooctene with
Reaction rates for oxidation of thioanisoles with various
kinds of p substituents were dependent on the electronic
effects of the substituents on the aromatic rings. Hammett
plots [log(k /k ) versus σ; Figure S1 in the Supporting In-
(μ-O)].[
a]
H
2
O
2
catalyzed by TPA
2
[{WO(O
2
)
2
}
2
X
H
formation] for the competitive oxidation of thioanisole and
p-substituted thioanisoles catalyzed by TPA ·I showed good
3
–
1
linearity. The negative ρ value (–0.72) indicates formation
of the electrophilic active oxygen species. Similar negative ρ
Entry
Additive
HClO
HNO
Zn(ClO
Cd(NO
R
0
[mmmin ]
Yield [%]
1
2
3
4
5
6
7
8
9
0
1
2
3
4
4
1.42Ϯ0.18
1.20Ϯ0.02
0.60Ϯ0.07
0.50Ϯ0.04
0.48Ϯ0.05
0.47Ϯ0.09
0.43Ϯ0.04
0.26Ϯ0.03
0.23Ϯ0.02
0.05Ϯ0.01
0.03Ϯ0.01
0.03Ϯ0.01
0.02Ϯ0.01
0.02Ϯ0.01
47
43
20
20
18
14
17
4
9
2
1
1
values have been reported for the oxidation of aryl sulfides
3
0
4
)
2
·6H
2
O
O
with H
2
O
2
catalyzed by d -transition-metal peroxo com-
[
9i,10g,14]
3
)
2
·4H
2
plexes such as Mo, W, and Re.
Ni(NO
Fe(ClO
Zn(NO
3 2 2
) ·6H O
4
The electronic character of the active oxygen species in
)
2
·nH
·6H
2
O
O
2
O
TPA ·I was investigated by using thianthrene 5-oxide
3
3
)
2
)
2
)
2
(
SSO), which has been used as a probe of the electronic
Co(ClO
4
)
2
·6H
·4H
·6H
[
15]
Ca(NO
Mg(NO
3
2
O
character of an oxidant [Equation (2)]. The oxidation re-
1
1
1
1
1
3
2
2
O
NaClO
LiClO ·3H
4
4
2
O
KClO
none
4
1
1
[
a] Reaction conditions: TPA
tive (0.36 mm), cyclooctene (0.14 m), H
acetonitrile (7 mL), 305 K, 1 h. Yield (%) = epoxide [mol]/initial
cyclooctene [mol]ϫ100. R values were determined from the reac-
tion profiles at the low conversion (Յ10%) of cyclooctene and
2
[{WO(O
2
)
2
}
2
2
(μ-O)] (0.71 mm), addi-
2
O
(0.14 m), H O (0.56 m),
2
0
2 2
H O .
Eur. J. Inorg. Chem. 2013, 1943–1950
1944
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
action did not proceed without TPA ·I under the present 4). The 17O NMR spectrum of TPA [{WO(O ) } (μ-O)]
3
2
2 2 2
conditions. The XSO value [XSO = (nucleophilic oxidation)/ showed two signals at δ = 647 and 243 ppm with the respec-
(
total oxidation)] was 0.13 and lower than those (0.14–0.35) tive intensity ratio of 2.0:1.0. It has been reported for perox-
[9i]
17
of the other peroxotungstates. These results suggest that otungstates that O signals of terminal (W=O) and bridg-
TPA ·I has the most electrophilic active oxygen species.
ing (W–O–W) oxygen atoms are observed in the range of δ
3
[
16]
=
550–700 and 100–550 ppm, respectively.
In addition,
the ratio of the number of terminal oxygen atoms to that
Solution State of I During the Catalysis
of bridging oxygen atoms in TPA [{WO(O ) } (μ-O)] is 2:1.
2
2 2 2
Therefore, the δ = 647 and 243 ppm signals are assigned to
We have proposed that TPA ·I is a catalytically active
3
terminal and bridging oxygen atoms, respectively (Table 2,
species and the reaction of an alkene with TPA ·I is the
3
17
entry 1). The O NMR spectrum of TPA ·I showed two
3
rate-determining step on the basis of the following re-
signals at δ = 692 and 105 ppm with the respective intensity
ratio of 2.8Ϯ0.2:1.0, which are assignable to terminal and
[
12]
sults:
(i) kinetic studies on epoxidation of cyclooctene
with H O catalyzed by TPA ·I showed first-order depen-
2
2
3
dences of reaction rates on concentrations of TPA ·I (0.23–
3
Table 2. 17O NMR spectroscopic parameters of TPA
(μ-O)], TPA ·I, TPA [{WO(O (μ-O )], and IЈ.
[{WO-
2
1.44 mm) and cyclooctene (0.04–0.43 m; Figure S2 in the
(
O
2
)
2
}
2
3
2
2
)
2
}
2
2
Supporting Information). (ii) Reaction rates were not de-
Entry Compound Chemical shift [ppm]
pendent on concentrations of H O (0.04–0.43 m), and the
2
2
Terminal
Bridging
(W–O–W)
dependence was different from that of [{WO(O ) } -
2
2 2
(W=O)
2
– [9e]
(
μ-O)] .
(iii) The epoxidation rate of TPA ·I was sixty
3
1
2
3
4
TPA
TPA
TPA
IЈ
2
3
2
[{WO(O
·I
2
)
2
2
}
2
2
(μ-O)]
(μ-O )]
647
692
662
688
243
105
not observed
332
times larger than that of TPA [{WO(O ) } (μ-O)].
2
2 2 2
Figure 2 shows the dependence of reaction rates on con-
centrations of water (0.08–1.01 m). Reaction rates decreased
as the concentrations of water increased. The effects of
H O and water on the solution state of TPA ·I were inves-
[{WO(O
2
)
}
2
2
2
3
1
7
tigated with O NMR spectroscopy (Table 2, Figures 3 and
Figure 3. 17O NMR spectra of (a) TPA
3
TPA ·I (solvent: [D ]acetonitrile; 253 K. [W] = 0.12 m).
[{WO(O
)
2
}
2
(μ-O)] and (b)
2
2
3
Figure 2. (a) Dependence of the reaction rates on concentrations
of water. (b) Plots of 1/R against concentrations of water. Reaction
conditions: TPA ·I (0.36 mm), cyclooctene (0.14 m), H (0.14 m),
O (0.08–1.01 m), acetonitrile (7 mL), 305 K. Solid lines were cal-
culated with [Equation (6), see below] and Equation (S15) in the
Supporting Information.
Figure 4. 17O NMR spectra of TPA
2 2
H O (0.15 m) and water [(a) 0.05, (b) 0.25, (c) 0.5, (d) 0.75, and
(e) 1.0 m]. The respective O contents in water were (a) 13.5, (b)
13.3, (c) 12.0, (d) 14.0, and (e) 12.0%. The observed spectra in the
region of δ = 650–710 ppm were thoroughly reproduced by the sum
(dotted lines) of the deconvoluted signals (gray solid lines).
3
·I (0.03 m) in the presence of
0
1
7
3
2
O
2
H
2
Eur. J. Inorg. Chem. 2013, 1943–1950
1945
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
bridging oxygen atoms, respectively (Table 2, entry 2).[17] It I688/I692 and I332/I105 ratios were proportional to the con-
has been reported for tungsten-, molybdenum-, vanadium-, centrations of water, the slopes were almost the same as
and niobium-based polyoxo- and peroxometalates that each other, and the value extrapolated at [H O] = 0 mm was
2
oxo Ǟ hydroxo transformations result in δ = 142–306 ppm almost zero (Figure 5, b). Therefore, IЈ would reversibly be
1
7
[16,18]
–1
upfield shifts of O NMR spectroscopic signals.
The formed according to Equation (3) (K = 2.0Ϯ0.3 m ). The
1
17
upfield shift from δ = 243 to 105 ppm is in accordance with
the protonation of bridging oxygen atoms.
Effects of water (0.05–1.00 m) on the solution state of spect to [{WO(O ) } (μ-O)] ), thus showing that IЈ is not
TPA ·I were investigated in the presence of H O (0.15 m; the species formed by the reaction of [{WO(O ) } (μ-O)]
O NMR spectrum of TPA [{WO(O ) } (μ-O)] was not
changed upon addition of water (3.3–16.7 equiv. with re-
2
2 2 2
[19,20]
2–
2
2 2
2–
3
2
2
2 2 2
[
21]
Figure 4).
Upon addition of five equivalents of 36% with water. In addition, polynuclear peroxotungstates that
H O with respect to TPA ·I ([water] = 0.5 m), three new consist of three or more peroxotungstate dimer units are
2
2
3
1
7
O signals appeared at δ = 688, 662, and 332 ppm (Fig- unprecedented. These facts support the idea that IЈ likely
ure 4, c). Separately, it was confirmed that the δ = 662 ppm has a tetranuclear structure. The reaction rates increased as
signal was assignable to the terminal oxygen atoms in concentrations of I increased, whereas no correlation be-
2
–
[22]
[
{WO(O ) } (μ-O )] (Table 2, entry 3). With increase in tween the rates and concentrations of IЈ was observed (Fig-
2 2 2 2
concentrations of water (0.05–1.00 m), the integrated inten- ure 2, a and Figure S3 in the Supporting Information).
sities of the δ = 688 and 332 ppm signals (I688 and I332 Therefore, IЈ would be inactive for the present epoxid-
increased, whereas those (I692 and I105) of TPA ·I decreased ation.
)
[
24]
3
(Figure S3 in the Supporting Information). Plots of I662 and
I688 against I332 are shown in Figure 5 (a). I688 was pro-
portional to I332, whereas such a correlation was not ob-
served between I662 and I332. Therefore, the δ = 688 and
3
32 ppm signals are assignable to the identical species (IЈ;
Reaction Mechanism
Table 2, entry 4). The chemical shifts were in the range of
terminal and bridging oxygen atoms of peroxotungstates,
On the basis of all the results, the present epoxidation
of alkenes possibly proceeds as follows (Figure 6): first, the
alkene directly attacks the peroxo oxygen groups in I to
form the corresponding epoxide and subsequent peroxo
species (II) [Equation (4)]. Then, I is regenerated by the re-
action of II with H O [Equation (5)]. According to the
[
16]
respectively.
In addition, the slope (i.e., I688/I332) was
2
.6Ϯ0.4 and close to 2.[17] All the results suggest that IЈ
consists of the peroxotungstate dimer units. The sum of I688
and I692 was almost constant, and I688 increased and I
decreased as the concentrations of water increased.
692
[
23]
The
2
2
proposed reaction mechanism in Figure 6, the overall epox-
idation rate is approximated by Equation (6) (see the Sup-
porting Information). On the basis of the kinetic data (Fig-
ure 2, b), the k1 and K1 values were calculated to be
–
1
–1
–1
(
5.9Ϯ0.5)ϫ10 m min and (2.3Ϯ0.3) m , respectively.
–1
The K value was almost the same as that [(2.0Ϯ0.3) m ]
1
determined from the 17O NMR spectra. Dependences of
reaction rates on concentrations of I, H O , cyclooctene,
2
2
and water and the time course of epoxidation were fairly
well reproduced by the numerical integration of Equa-
tion (6) (shown by solid lines in Figure 2 and Figures S2
and S5 in the Supporting Information). These results sup-
port the proposed mechanism.
To investigate the proposed reaction mechanism, DFT
calculations were carried out at the B3LYP/LanL2DZ+6-
31+G(d,p) level. Reactivities of eight kinds of peroxo oxy-
gen atoms (O2–O5 and O7–O10) in I for the epoxidation
of ethylene in the gas phase were investigated. The transi-
tion-state structures and activation barriers were calculated
on the assumption that the alkene double bond directly at-
Figure 5. (a) Plots I668 and I688 against I332. (b) Plots of I688/I692
2
and I332/I105 against concentrations of water. Slope was 0.002 (R
0.95).
=
Eur. J. Inorg. Chem. 2013, 1943–1950
1946
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
–1
was calculated to be exothermic by 38 kJmol . The com-
putational results support the idea that the oxygen-transfer
step is the rate-determining step.
Conclusion
In conclusion, we have investigated the effects of addi-
tives on the epoxidation of cyclooctene with H O catalyzed
2
2
by TPA [{WO(O ) } (μ-O)]. Among the additives tested,
2
2 2 2
HClO is the most effective. Upon addition of 0.5 equiv. of
4
HClO with respect to TPA [{WO(O ) } (μ-O)], a tetranu-
4
2
2 2 2
clear peroxotungstate TPA ·I is formed in situ. The low X
3
SO
Figure 6. Proposed reaction mechanism for epoxidation of alkenes
with H catalyzed by I.
value for the TPA ·I-catalyzed oxidation of SSO relative to
2
O
2
3
those of the other peroxotungstates shows the generation of
a strongly electrophilic oxygen species on I. The kinetic,
tacks the peroxo oxygen groups in a spiro fashion (i.e., the mechanistic, and spectroscopic investigation show that I is
CCO plane is almost orthogonal to the WOO plane). The the active species for the present epoxidation and that oxy-
transition-state structures and the corresponding activation gen transfer from I to a C=C double bond is the rate-de-
barrier are shown in Figure S6 in the Supporting Infor- termining step.
mation. The calculated activation barriers of an ethylene
–
1
attack at O10 (TS1, 108 kJmol ) was the lowest among
those of ethylene attacks at the other oxygen atoms (115–
Experimental Section
–
1
130 kJmol ), thereby suggesting that an alkene attack at
Materials: Acetonitrile (Kanto Chemical) was purified by the Ulti-
mate Solvent System (GlassContour Company) prior to use.
Substrates were purified according to the reported procedures.
O10 is the most favorable.
[
27]
28]
The energies of the reaction steps were calculated accord-
ing to Figure 6 by taking into account the solvation in
acetonitrile using the conductor-like polarizable continuum
model (CPCM) with the parameters of the universal force
field (UFF) model. The results are summarized in Figure 7.
The formation of a subsequent peroxo species (II) and eth-
[
H
2
O
2
(30% aqueous solution, Kanto Chemical), H
2
WO
4
(Wako
(70%
Chemical), TPA·OH (1 m aqueous solution, Aldrich), HNO
aqueous solution, Kanto Chemical), and metal salts (Kanto Chem-
ical and Wako Chemical) were used as received. A solution of 94%
3
2 2
aqueous H O was prepared by concentration of 30% aqueous
ylene oxide by the reaction of ethylene with I was calculated H O . Deuterated solvents (CD CN, CDCl , and D O) were pur-
2
2
3
3
2
–
1
17
17
18
to be exothermic by 173 kJmol . The activation barrier for chased from ACROS and used as received. H2 O ( O, 20.2%; O,
the oxygen-transfer reaction (TS1–CH CN) was calculated 67.2%) was purchased from ISOTEC and used as received.
3
–
1 [25,26]
to be 85 kJmol .
The reaction of II with H O to
2 2
Instruments: IR spectra were measured with a Jasco FT/IR-460
spectrometer Plus using KCl disks. UV/Vis spectra were recorded
with a Jasco V-570 spectrometer. NMR spectra were recorded with
form I and water accomplishes the catalytic cycle. This step
1
13
a
6
JEOL JNM-EX-270 spectrometer ( H 270.0 MHz,
C
1
7
183
7.80 MHz, O 36.45 MHz,
W 11.20 MHz) by using 5 mm
tubes (for H and 13C) or 10 mm tubes (for 17O and W). Chemi-
1
183
3 4
cal shifts were reported in ppm downfield from Si(CH ) (solvent,
1
13
17
CDCl
3
) for H and C NMR spectra, D
WO (solvent, D
spectra. O NMR spectra were measured upon addition of H
2
O (solvent, D
2
O) for
O
1
83
NMR spectra, and 2 m Na
2
4
2
O) for
W NMR
1
7
17
2
O
1
7
(
O content: 20.2%) into the solution that contained peroxotungs-
tates. The integrated intensities were normalized by O contents.
Only O signals assignable to terminal and bridging oxygen atoms
could be detected, thus showing that peroxo oxygen atoms are not
1
7
1
7
1
7
exchanged by oxygen-17 in H
2
O but rather terminal and bridging
oxygen atoms, in accord with the literature.[
16,29]
GC analyses were
carried out with a Shimadzu GC-2014 with a flame ionization de-
tector equipped with a DB-WAX extended temperature range
(
ETR) capillary column (internal diameter: 0.25 mm, length: 30 m)
or a TC-1 capillary column (internal diameter: 0.25 mm, length:
0 m). HPLC analyses were performed with an Agilent 1100 series
3
LC with a UV/Vis detector using a CAPCELL PAK MG C18 re-
verse-phase column (5 μmϫ3 mmϫ250 mm, Shiseido Fine Chem-
icals). Mass spectra were recorded with a Shimadzu GC–MS-
QP2010 equipped with a TC-5HT capillary column at an ionization
voltage of 70 eV.
Figure 7. Calculated energy diagram of epoxidation of ethylene by
I in acetonitrile (energies and lengths are in kJmol–1 and Å, respec-
tively). Gray, red, black, and light-blue balls represent tungsten,
oxygen, carbon, and hydrogen atoms, respectively.
Eur. J. Inorg. Chem. 2013, 1943–1950
1947
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Synthesis and Characterization of TPA
3
·I:
A
TPA salt of
cluded. All calculations were performed with the Gaussian 09 pro-
gram package.
3–
[12]
[35]
[
H{W
2
O
2
(O
2
)
4
(μ-O)}
2
]
was
synthesized
as
follows:
TPA [{WO(O
2
2
)
2
}
2
(μ-O)] (53 mg, 58 μmol) and 70% HNO
3
Supporting Information (see footnote on the first page of this arti-
cle): Kinetic derivation, Table S1, and Figures S1–S7. Molecular
structures of all peroxotungstates under examination are shown in
Figure S7.
(100 μmol) were dissolved in acetonitrile (0.5 mL) that contained
30% aqueous H (0.05 mL). Diethyl ether (0.5 mL) was added
2 2
O
to the resulting solution. Colorless crystals suitable for X-ray crys-
tallographic analysis were obtained by vapor diffusion of diethyl
ether into the solution at 277 K. Yield: 33 mg (69%). 1 O NMR
7
(
36.45 MHz, CD
3 2
CN, 253 K, D O): δ = 692.7 (Δν1/2 = 138 Hz),
Acknowledgments
1
03.6 (Δν1/2 = 621 Hz) ppm with the respective intensity ratio of
2.8Ϯ0.2):1.0. 1 W NMR (11.20 MHz, CD
83
(
3 2 4
CN, 253 K, Na WO ):
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 Pro-
gram) and grants-in-aid for JSPS Fellows and Scientific Research
from the Ministry of Education, Culture, Science, Sports, and Tech-
nology of Japan.
δ = –565.6 (Δν1/2 = 1.6 Hz) ppm. IR (KCl): ν˜ = 1038, 981, 967,
–1
8
1
{
4
44, 755, 707, 670, 642, 585, 573, 511, 377 cm . Raman: ν˜ = 1136,
–1
105, 1036, 973, 859, 580, 526, 309 cm .
TPA [H{W (O (μ-O)} ]}: calcd. C 26.25, H 5.20, N 2.55, W
4.64; found C 26.01, H 5.31, N 2.53, W 44.45.
36 85 3 22 4
C H N O W
3
2
O
2
2
)
4
2
2 2
Typical Procedure for Catalytic Epoxidation of Alkenes with H O :
A catalytic reaction was carried out with a glass tube (30 mL) that
contained a magnetic stirrer bar. The solvent, substrate, oxidant,
and additive were added to the reaction vessel. The reaction was
initiated by the addition of the catalyst, and the reaction solution
was periodically analyzed by GC. All products were known com-
pounds and identified by comparison of their retention times, mass
spectra, and NMR spectra with those of authentic samples.
[1] Modern Oxidation Methods (Ed.: J.-E. Bäckvall), Wiley-VCH,
Weinheim, Germany, 2004.
[
2] a) Mechanisms in Homogeneous and Heterogeneous Epoxidation
Catalysis (Ed.: S. T. Oyama), Elsevier, Amsterdam, 2008; b)
B. S. Lane, K. Burgess, Chem. Rev. 2003, 103, 2457–2473.
3] a) J. Rudolph, K. L. Reddy, J. P. Chiang, K. B. Sharpless, J.
Am. Chem. Soc. 1997, 119, 6189–6190; b) W. A. Herrmann,
F. E. Kühn, M. K. Mattner, G. R. J. Artus, M. R. Geisberger,
J. D. G. Correia, J. Organomet. Chem. 1997, 538, 203–209; c)
H. Rudler, J. R. Gregorio, B. Denise, J.-M. Brégeault, A. Delof-
fre, J. Mol. Catal. A 1998, 133, 255–265; d) W. A. Herrmann,
R. M. Kratzer, H. Ding, W. R. Thiel, H. Glas, J. Organomet.
Chem. 1998, 555, 293–295; e) H. Adolfsson, A. Converso, K. B.
Sharpless, Tetrahedron Lett. 1999, 40, 3991–3994; f) H. Adolfs-
son, C. Copéret, J. P. Chiang, A. K. Yudin, J. Org. Chem. 2000,
65, 8651–8658; g) C. Copéret, H. Adolfsson, K. B. Sharpless,
Chem. Commun. 1997, 1565–1566.
[
Procedure for Oxidation of SSO: Acetonitrile (5 mL), SSO
(
200 μmol), 94% H
dard) were added to the reaction vessel, and the solution was co-
oled to 273 K. The reaction was initiated by the addition of TPA ·I
2.5 μmol). The oxidation products (SOSO = thianthrene 5,10-di-
oxide, SSO = thianthrene 5,5-dioxide, and SOSO = thianthrene
,5,10-trioxide) were analyzed quantitatively by HPLC: eluent,
2 2
O (50 μmol), and naphthalene (internal stan-
3
(
2
2
5
–1
3 3 2
CH OH/CH CN/H O = 60:15:25; flow rate 0.4 mLmin , column
temperature 303 K, detection at λ = 254 nm. The reaction condi-
tions (e.g., concentrations of substrate, substrate/oxidant ratios,
and reaction temperatures) were controlled to minimize overoxid-
[
4] a) B. S. Lane, K. Burgess, J. Am. Chem. Soc. 2001, 123, 2933–
2934; b) B. S. Lane, M. Vogt, V. J. DeRose, K. Burgess, J. Am.
Chem. Soc. 2002, 124, 11946–11954; c) P. L. Anelli, S. Banfi,
F. Montanari, S. Quici, J. Chem. Soc., Chem. Commun. 1989,
2
ation to SOSO and then to estimate the true electronic nature
[
30]
of the oxidant.
following equation reported:[
tal oxidation) = (SSO + SOSO
The XSO value was calculated according to the
7
79–780; d) A. M. d’A. R. Gonsalves, A. C. Serra, J. Mol. Ca-
30]
XSO = (nucleophilic oxidation)/(to-
tal. A 2001, 168, 25–32; e) R. Krishnan, S. Vancheesan, J. Mol.
Catal. A 2000, 157, 15–24; f) G. B. Shul’pin, G. Süss-Fink, L. S.
Shul’pina, J. Mol. Catal. A 2001, 170, 17–34.
2
2
)/(SSO + SOSO + 2SOSO ).
2
2
Kinetic Studies: A glass tube (15 mL) was used as a reactor for
kinetic studies on epoxidation of cyclooctene with H catalyzed
by TPA ·I. Concentrations of water (0.56 m) were fixed whereas the
concentrations of H (0.04–0.43 m) were varied by mixing 78%
and water (Figure S2c in the Supporting Information). The
reaction was periodically monitored by GC, and the remaining
was decomposed at 273 K by the addition of Ru(OH) /Al
values were determined from the re-
action profiles at low conversions (Ͻ10%) of both cyclooctene and
[5] a) W. Nam, H. J. Lee, S.-Y. Oh, C. Kim, H. G. Jang, J. Inorg.
Biochem. 2000, 80, 219–225; b) M. C. White, A. G. Doyle, E. N.
Jacobsen, J. Am. Chem. Soc. 2001, 123, 7194–7195.
6] a) P. Jiménez-Lozano, I. D. Ivanchikova, O. A. Kholdeeva,
J. M. Poblet, J. J. Carbó, Chem. Commun. 2012, 48, 9266–9268;
b) H. Gao, G. Lu, J. Suo, S. Li, Appl. Catal. A 1996, 138,
2
O
2
3
[
2
O
2
2 2
H O
2
7–38; c) C. N. Kato, S. Negishi, K. Yoshida, K. Hayashi, K.
H
2
O
2
x
2 3
O
Nomiya, Appl. Catal. A 2005, 292, 97–104; d) B. Notari, Adv.
Catal. 1996, 41, 253–334.
[
31]
before the GC analysis.
R
0
[
7] a) Y. Ishii, K. Yamawaki, T. Ura, H. Yamada, T. Yoshida, M.
Ogawa, J. Org. Chem. 1988, 53, 3587–3593; b) K. Sato, M.
Aoki, M. Ogawa, T. Hashimoto, D. Panyella, R. Noyori, Bull.
Chem. Soc. Jpn. 1997, 70, 905–915.
2 2
H O .
Quantum Chemical Calculations: DFT calculations were carried
out at the B3LYP level theory[ [6-31+G(d,p) basis sets for H, C,
O, Si, P, and S atoms and the double-ξ quality basis sets with effec-
tive core potentials proposed by Hay and Wadt[ for W, As, and
Se atoms] by using the CPCM with the parameters of the universal
32]
[8] a) M. T. Pope, Heteropoly and Isopoly Oxometalates, Springer,
Berlin, 1983; b) C. L. Hill, C. M. Prosser-McCartha, Coord.
Chem. Rev. 1995, 143, 407–455; c) T. Okuhara, N. Mizuno, M.
Misono, Adv. Catal. 1996, 41, 113–252; d) R. Neumann, Prog.
Inorg. Chem. 1998, 47, 317–370; e) Chem. Rev. 1998, 98, 1–389;
thematic issue on polyoxometalates (Ed.: C. L. Hill); f) I. V.
Kozhevnikov, Catalysts for Fine Chemical Synthesis vol. 2, Ca-
talysis by Polyoxometalates, Wiley, Chichester, UK, 2002; g)
M. T. Pope, in: Comprehensive Coordination Chemistry II, Vol.
33]
[
34]
force field (UFF).
Entire structures of peroxotungstates were
used as models in the calculations. Ethylene was used as a model
substrate. All geometries were optimized without symmetry restric-
tions. Transition-state structures were searched by numerically esti-
mating the matrix of second-order energy derivatives at every opti-
mization step and by requiring exactly one eigenvalue of this matrix
to be negative. The zero-point vibrational energies were not in-
4
(Eds.: J. A. McCleverty, T. J. Meyer), Elsevier Pergamon, Am-
sterdam, 2004, pp. 635–678; h) C. L. Hill, in: Comprehensive
Coordination Chemistry II, vol. 4 (Eds.: J. A. McCleverty, T. J.
Eur. J. Inorg. Chem. 2013, 1943–1950
1948
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Meyer), Elsevier Science, New York, 2004, pp. 679–759; i) N.
Mizuno, K. Kamata, K. Yamaguchi, in: Surface and Nanomo-
lecular Catalysis (Ed.: R. Richards), Taylor and Francis, New
York, 2006, pp. 463–492; j) N. Mizuno, K. Kamata, S. Uchida,
K. Yamaguchi, in: Modern Heterogeneous Oxidation Catalysis
[15] a) W. Adam, D. Golsch, Chem. Ber. 1994, 127, 1111–1113; b)
W. Adam, D. Golsch, J. Org. Chem. 1997, 62, 115–119; c) M.
Bonchio, V. Conte, M. Assunta, D. Conciliis, F. D. Furia, F. P.
Ballistreri, G. A. Tomaselli, R. M. Toscano, J. Org. Chem.
1995, 60, 4475–4480; d) W. Adam, C. M. Mitchell, C. R. Saha-
Möller, T. Selvam, O. Weichold, J. Mol. Catal. A 2000, 154,
251–255; e) K. Sato, M. Hyodo, M. Aoki, X.-Q. Zheng, R.
Noyori, Tetrahedron 2001, 57, 2469–2476; f) W. A. Herrmann,
R. W. Fischer, J. D. G. Correia, J. Mol. Catal. 1994, 94, 213–
223.
(
1
Ed.: N. Mizuno), Wiley-VCH, Weinheim, Germany, 2009, pp.
85–216; k) N. Mizuno, K. Kamata, K. Yamaguchi, Top. Ca-
tal. 2010, 53, 876–893.
[
9] a) Y. Ishii, T. Yoshida, K. Yamawaki, M. Ogawa, J. Org. Chem.
1988, 53, 5549–5552; b) C. Venturello, R. D’Aloisio, J. C. J.
Bart, M. Ricci, J. Mol. Catal. 1985, 32, 107–110; c) D. C. Dun- [16] O. W. Howarth, Dalton Trans. 2004, 476–481.
can, R. C. Chambers, E. Hecht, C. L. Hill, J. Am. Chem. Soc.
[17] The slight discrepancies between the observed and theoretical
995, 117, 681–691; d) K. Kamata, K. Yamaguchi, N. Mizuno,
1
ratios (2:1) are probably explained by underestimation of the
broad δ = 105 and 332 ppm signal intensities.
Chem. Eur. J. 2004, 10, 4728–4734; e) K. Kamata, S. Kuzuya,
K. Uehara, S. Yamaguchi, N. Mizuno, Inorg. Chem. 2007, 46,
[18] a) L. Pettersson, I. Andersson, F. Taube, I. Toth, M. Hashim-
oto, O. W. Howarth, Dalton Trans. 2003, 146–152; b) R. I.
Maksimovskaya, G. M. Maksimov, Inorg. Chem. 2011, 50,
4725–4731; c) W. G. Klemperer, W. Shum, J. Am. Chem. Soc.
1978, 100, 4891–4893; d) J. R. Black, M. Nyman, W. H. Casey,
J. Am. Chem. Soc. 2006, 128, 14712–14720; e) M. Pohl, R. G.
Finke, Organometallics 1993, 12, 1453–1457.
3
768–3774; f) J.-M. Brégeault, M. Vennat, L. Salles, J.-Y.
Piquemal, Y. Mahha, E. Briot, P. C. Bakala, A. Atlamsani, R.
Thouvenot, J. Mol. Catal. A 2006, 250, 177–189; g) K. Kam-
ata, T. Hirano, S. Kuzuya, N. Mizuno, J. Am. Chem. Soc. 2009,
1
31, 6997–7004; h) K. Kamata, T. Hirano, N. Mizuno, Chem.
Commun. 2009, 3958–3960; i) K. Kamata, T. Hirano, R. Ishim-
oto, N. Mizuno, Dalton Trans. 2010, 39, 5509–5518; j) K. Kam-
ata, R. Ishimoto, T. Hirano, S. Kuzuya, K. Uehara, N. Mizuno,
Inorg. Chem. 2010, 49, 2471–2478.
1
7
[19] The O signal broadening of bridging oxygen atoms in I is
likely owing to proton exchange between two [{WO(O ) } (μ-
2
2 2
2
–
[20]
O)] units; see ref.
[
10] a) K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S. Hi-
kichi, N. Mizuno, Science 2003, 300, 964–966; b) K. Kamata,
Y. Nakagawa, K. Yamaguchi, N. Mizuno, J. Catal. 2004, 224,
[20] L. Kimtys, V. Balevi cˇ ius, D. W. Aksnes, J. Mol. Struct. 1992,
270, 161–171.
[21] Upon addition of water ([TPA ·I] = 0.03 m, [water] = 0.25 m)
3
1
7
2
24–228; c) K. Kamata, M. Kotani, K. Yamaguchi, S. Hikichi,
2 2
in the absence of H O , the O NMR spectroscopic signal of
1
7
N. Mizuno, Chem. Eur. J. 2007, 13, 639–648; d) D. G. Musaev,
K. Morokuma, Y. V. Geletii, C. L. Hill, Inorg. Chem. 2004, 43,
IЈ appeared in addition to several unidentified O signals. The
intensities of the unidentified signals increased greatly with
time, oxygen was evolved, and the signals of I and IЈ disap-
peared, thereby suggesting the self-decomposition of I. It was
also confirmed by UV/Vis spectroscopy that the intensity of
the 243 nm absorption band characteristic of peroxotungstates
decreased with time in the absence of H O .
7
702–7708; e) R. Prabhakar, K. Morokuma, C. L. Hill, D. G.
Musaev, Inorg. Chem. 2006, 45, 5703–5709; f) A. Sartorel, M.
Carraro, A. Bagno, G. Scorrano, M. Bonchio, Angew. Chem.
2
007, 119, 3319; Angew. Chem. Int. Ed. 2007, 46, 3255–3258;
g) M. Carraro, L. Sandei, A. Sartorel, G. Scorrano, M. Bon-
2
2
chio, Org. Lett. 2006, 8, 3671–3674; h) T. D. Phan, M. A. [22] Upon addition of 78% H O (2.5 equiv. with respect to
2
2
2
–
Kinch, J. E. Barker, T. Ren, Tetrahedron Lett. 2005, 46, 397–
00; i) A. Yoshida, M. Yoshimura, K. Uehara, S. Hikichi, N.
Mizuno, Angew. Chem. 2006, 118, 1990; Angew. Chem. Int. Ed.
006, 45, 1956–1960; j) R. Ishimoto, K. Kamata, N. Mizuno,
Angew. Chem. 2009, 121, 9062; Angew. Chem. Int. Ed. 2009,
8, 8900–8904.
11] a) R. Neumann, M. Gara, J. Am. Chem. Soc. 1995, 117, 5066–
074; b) M. Bösing, A. Nöh, I. Loose, B. Krebs, J. Am. Chem.
Soc. 1998, 120, 7252–7259; c) M. D. Ritorto, T. M. Anderson, [23] I662 was not much changed, and the sum of the five signal
2 2 2 2 2 2
[{WO(O ) } (μ-O)] ), the signal intensities of [{WO(O ) } (μ-
2
–
4
O)] were much decreased and a new signal appeared at δ =
2
–
662 ppm. It has been reported that [{WO(O ) } (μ-O )] is
2
2
2
2
2
–
2
2 2 2 2 2
generated by the reaction of [{WO(O ) } (μ-O)] with H O ;
2
–
in addition, [{WO(O ) } (μ-O )] has two terminal oxygen
2
2
2
2
4
atoms and no bridging oxygen atom. Therefore, the δ =
[
662 ppm signal is assignable to terminal oxygen atoms in
2
–
5
2 2 2 2
[{WO(O ) } (μ-O )] .
W. A. Neiwert, C. L. Hill, Inorg. Chem. 2004, 43, 44–49; d) N.
Mizuno, C. Nozaki, I. Kiyoto, M. Misono, J. Am. Chem. Soc.
intensities remained almost constant.
1
7
[24] The O signal of bridging oxygen atoms in IЈ was observed at
lower magnetic field than that in I. Therefore, water possibly
works as a proton acceptor to form an inactive nonprotonated
tetranuclear peroxotungstate (IЈ). It has been reported that re-
1
998, 120, 9267–9272; e) Y. Nakagawa, K. Kamata, M. Kotani,
K. Yamaguchi, N. Mizuno, Angew. Chem. 2005, 117, 5266; An-
gew. Chem. Int. Ed. 2005, 44, 5136–5141; f) K. Kamata, K.
Yonehara, Y. Nakagawa, K. Uehara, N. Mizuno, Nat. Chem.
action rates increase with increase in the W–O bond lengths in
di- and tetranuclear peroxotungstates.[
9g,12]
DFT calculations
2
010, 2, 478–483; g) K. Kamata, K. Sugahara, K. Yonehara,
R. Ishimoto, N. Mizuno, Chem. Eur. J. 2011, 17, 7549–7559;
h) K. Kamata, T. Yamaura, N. Mizuno, Angew. Chem. Int. Ed.
2 2
showed that the W–OW bond lengths (1.93 Å) of [{W O -
4
–
(O ) (μ-O)} ] are much shorter than those (1.99–2.06 Å) of
2
4
2
2
012, 51, 7275–7279.
12] R. Ishimoto, K. Kamata, N. Mizuno, Angew. Chem. Int. Ed.
012, 51, 4662–4665.
other peroxotungstates including I, thereby supporting the idea
that IЈ is inactive for epoxidation (Figure S4 in the Supporting
Information).
[
[
2
–
1
13] a) W. Adam, A. Corma, H. García, O. Weichold, J. Catal. [25] This calculated activation barrier (85 kJmol ) is in the range
–
1
of those (53–99 kJmol ) reported for titanium-substituted
2
000, 196, 339–344; b) W. Adam, C. M. Mitchell, C. R. Saha-
POMs; see ref.[
6a,26]
Möller, Eur. J. Org. Chem. 1999, 785–790; c) G. Bellucci, G.
Berti, M. Ferretti, G. Igrosso, E. Mastrorilli, J. Org. Chem. [26] N. S. Antonova, J. J. Carbó, U. Korz, O. A. Kholdeeva, J. M.
1
978, 43, 422–428; d) R. W. Murray, M. Singh, B. L. Williams,
Poblet, J. Am. Chem. Soc. 2010, 132, 7488–7497.
[27] A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen,
F. J. Timmers, Organometallics 1996, 15, 1518–1520.
H. M. Moncrieff, J. Org. Chem. 1996, 61, 1830–1841.
[
14] a) A. Arcoria, F. P. Ballistreri, G. A. Tomaselli, J. Mol. Catal.
[28] W. L. F. Armarego, C. L. L. Chai, Purification of Laboratory
1983, 18, 177–188; b) K. A. Vassell, J. H. Espenson, Inorg.
Chem. 1994, 33, 5491–5498; c) A. M. Al-Ajlouni, J. H. Espen-
son, J. Am. Chem. Soc. 1995, 117, 9243–9250; d) G. Du, J. H.
Espenson, Inorg. Chem. 2005, 44, 2465–2471; e) F. E. Kühn,
A. M. Santos, W. A. Herrmann, Dalton Trans. 2005, 2483–
Chemicals, 5th ed., Butterworth-Heinemann, Oxford, 2003.
[29] M. Postel, C. Brevard, H. Arzoumanian, J. G. Riess, J. Am.
Chem. Soc. 1983, 105, 4922–4926.
[30] W. Adam, D. Golsch, F. C. Görth, Chem. Eur. J. 1996, 2, 255–
2491.
258.
Eur. J. Inorg. Chem. 2013, 1943–1950
1949
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[
31] a) K. Yamaguchi, N. Mizuno, Angew. Chem. 2002, 114, 4720;
Angew. Chem. Int. Ed. 2002, 41, 4538–4542; b) K. Yamaguchi,
N. Mizuno, Chem. Eur. J. 2003, 9, 4353–4361.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-
erov, T. Keith, R. Kobayashi, J. 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. Strat-
mann, 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,
D. J. Fox, Gaussian 09, rev. B.01, Gaussian, Inc., Wallingford,
CT, 2010.
[
[
[
32] A. D. Becke, J. Chem. Phys. 1993, 98, 1372–1377.
33] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270–283.
34] A. K. Rappé, C. J. Casewit, K. S. Colwell, W. A. Goddard III,
W. M. Skiff, J. Am. Chem. Soc. 1992, 114, 10024–10035.
35] 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. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M.
[
Received: September 12, 2012
Published Online: December 4, 2012
Eur. J. Inorg. Chem. 2013, 1943–1950
1950
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim