Reactivity patterns of O2, CO2, carboxylic acids, and triflic acid with molybdenum silyl hydrido complexes bearing polydentate phosphinoalkyl-silyl ligands: Pronounced effects of silyl ligands on reactions

Article abstract of DOI:10.1021/om200646c

The reactivity patterns of a series of molybdenum silyl hydrido complexes, [MoH₃{[Ph₂PCH₂CH₂P(Ph)C ₆H₄-o]₂(R)Si-P,P,P,P,Si}] (2a, R = Ph; 2b, R = C₆F₅; 2c, R = 4-Me₂NC₆H₄; 2d, R = cyclohexyl; 2e, R = n-C₆H₁₃), toward electrophilic materials, including dioxygen, carbon dioxide, carboxylic acids, and triflic acid, have been explored. The influence of substituents directly attached to the Si atom in the quintuply chelating diphosphine-silyl ligands during the course of the reactions has been studied in detail. Complexes 2a-c react readily with dioxygen at ambient temperature, yielding η²-O₂ complexes 3a-c, in which the dioxygen ligand ligates at the site opposite to the Si atom. Single-crystal X-ray studies have established that the Mo-O distance of 3a, which contains a phenyl-substituted Si fragment, is significantly longer than that found in common η²-O₂ complexes, in which the sizes of the central atoms are nearly equal to that of Mo as a consequence of the strong trans influence of the silyl ligand. In contrast, the replacement of the phenyl group on a Si fragment by an electron-withdrawing perfluoro phenyl group, as in 3b, gives rise to a dramatic decrease in the trans influence. Treatment of 2a-c with 2 equiv of carboxylic acids, such as formic acid, acetic acid, and benzoic acid, readily leads to the formation of carboxylato molybdenum complexes, in which the carboxylate ligand coordinates to the metal in a monodentate mode. The resulting formato complex 4a1 can also be derived from the insertion of CO ₂ into the Mo-H bond of 2a. In addition, complexes 2a,c,e mediate the catalytic carbon dioxide hydrogenation. The activity of the catalysts depends dramatically on the electronic effects of the substituents attached to the Si atom; electron-donating substituents lead to an increase in the catalytic activity. Reactions of complexes 2a,d with triflic acid yield highly labile triflato complexes, which can be further treated with aqueous NaOH to give hydroxo complexes 5a,d, respectively. The molecular structure of 5d has been determined by single-crystal X-ray diffraction. The reactivities of 2a-e are discussed in comparison with those of the parent complex [MoH ₄(dppe)₂] (1; dppe = Ph₂PCH₂CH ₂PPh₂).

Full text of DOI:10.1021/om200646c

Article
pubs.acs.org/Organometallics
Reactivity Patterns of O2, CO2, Carboxylic Acids, and Triflic Acid with
Molybdenum Silyl Hydrido Complexes Bearing Polydentate
Phosphinoalkyl−Silyl Ligands: Pronounced Effects of Silyl Ligands on
Reactions
Makoto Minato,*,† Da-Yang Zhou,†,§ Ken-ichiro Sumiura,† Yuki Oshima,† Shigeki Mine,† Takashi Ito,†
Masaki Kakeya,† Kei Hoshino,† Takahiro Asaeda,† Takushi Nakada,† and Kohtaro Osakada‡
†Department of Materials Chemistry, Graduate School of Engineering, Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku,
Yokohama, 240-8501, Japan
‡Research Laboratory of Resources Utilization, Tokyo Institute of Technology, 4259, Nagatsuta, Midori-ku, Yokohama, 226-8503,
Japan
S
* Supporting Information
ABSTRACT: The reactivity patterns of a series of molybde-
num silyl hydrido complexes, [MoH3{[Ph2PCH2CH2P(Ph)-
C6H4-o]2(R)Si-P,P,P,P,Si}] (2a, R = Ph; 2b, R = C6F5; 2c, R =
4-Me2NC6H4; 2d, R = cyclohexyl; 2e, R = n-C6H13), toward
electrophilic materials, including dioxygen, carbon dioxide,
carboxylic acids, and triflic acid, have been explored. The
influence of substituents directly attached to the Si atom in the
quintuply chelating diphosphine−silyl ligands during the
course of the reactions has been studied in detail. Complexes
2a−c react readily with dioxygen at ambient temperature,
yielding η2-O2 complexes 3a−c, in which the dioxygen ligand
ligates at the site opposite to the Si atom. Single-crystal X-ray studies have established that the Mo−O distance of 3a, which
contains a phenyl-substituted Si fragment, is significantly longer than that found in common η2-O2 complexes, in which the sizes
of the central atoms are nearly equal to that of Mo as a consequence of the strong trans influence of the silyl ligand. In contrast,
the replacement of the phenyl group on a Si fragment by an electron-withdrawing perfluoro phenyl group, as in 3b, gives rise to a
dramatic decrease in the trans influence. Treatment of 2a−c with 2 equiv of carboxylic acids, such as formic acid, acetic acid, and
benzoic acid, readily leads to the formation of carboxylato molybdenum complexes, in which the carboxylate ligand coordinates
to the metal in a monodentate mode. The resulting formato complex 4a1 can also be derived from the insertion of CO2 into the
Mo−H bond of 2a. In addition, complexes 2a,c,e mediate the catalytic carbon dioxide hydrogenation. The activity of the catalysts
depends dramatically on the electronic effects of the substituents attached to the Si atom; electron-donating substituents lead to
an increase in the catalytic activity. Reactions of complexes 2a,d with triflic acid yield highly labile triflato complexes, which can
be further treated with aqueous NaOH to give hydroxo complexes 5a,d, respectively. The molecular structure of 5d has been
determined by single-crystal X-ray diffraction. The reactivities of 2a−e are discussed in comparison with those of the parent
complex [MoH4(dppe)2] (1; dppe = Ph2PCH2CH2PPh2).
compound I. Even in the case of a secondary silane system, a
similar reaction occurred to give the hydrido complex II with a
tridentate ligand comprised of a P2Si framework (Chart 1).
Multidentate ligand systems containing Si as a donor atom
have received increasing attention because the incorporation of
a silyl group as part of the ligand framework could result in a
number of unique properties for a metal.3,4 These compounds
are of interest for several reasons; among them, one is that silyl
ligands are expected to exhibit an exceptionally strong trans
effect due to their remarkable σ-donor and high trans-
INTRODUCTION
■
Transition-metal complexes bearing silyl ligands continue to
play an important role in organotransition-metal chemistry.1
We previously reported the unexpected formation of novel
m o l y b d e n u m s i l y l h y d r i d o c o m p l e x e s ,
[MoH3{[Ph2PCH2CH2P(Ph)C6H4-o]2(R)Si-P,P,P,P,Si}] (2a,
R = Ph; 2b, R = C6F5; 2c, R = 4-Me2NC6H4; 2d, R =
cyclohexyl; 2e, R = n-C6H13), with a pentadentate ligand
composed of a P2SiP2 framework, which can be readily made by
the reaction of [MoH4(dppe)2] (1; dppe = Ph2PCH2CH2PPh2)
with RSiH3 in refluxing toluene (Scheme 1).2 We found that
this type of reaction was not limited to primary silanes. The
reaction of PhGeH3 with 1 afforded the related germyl
Received: July 19, 2011
Published: July 9, 2012
© 2012 American Chemical Society
4941
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949

Organometallics
Article
corresponding adducts (R = Ph (3a), C6F5 (3b), 4-Me2NC6H4
(3c)), which can be isolated as crystalline solids (Scheme 2). In
Scheme 1
Scheme 2
Chart 1
Table 1. Preparation of (η2-O) Complexes 3a−c
compd substrate complex (R)/amt, mmol solvent/amt, mL yield, %
3a
3b
3c
2a (Ph)/0.217
THF/22
87
89
82
2b (C6F5)/0.092
toluene/10
toluene/10
2c (4-Me2NC6H4)/0.096
Table 1, we summarize the reaction conditions. The evolution
of a significant amount of dihydrogen was monitored using
GLC when the reaction of 2a with dioxygen was carried out in
a sealed system.
influencing ability.5 In this context, we suspected that
complexes 2a−e may be excellent candidates for conducting
a detailed investigation into the trans influence of silyl ligands.
Indeed, the presence of the labile hydride ligands in the
coordination sphere of molybdenum provides us with the
ability to open up a site opposite to the Si atom, under mild
conditions, for an incoming ligand via the reductive elimination
of H2. Our recent work has demonstrated that complexes 2a−c
react rapidly with isocyanides to give isocyanido complexes in
which the isocyanido group always ligates at the site opposite to
the Si atom.6 Furthermore, we have revealed that the trans-
influencing ability imparted by the Si fragment in these
isocyanido complexes is intimately dependent on the donor
properties of the substituent attached to the Si atom.
In complexes 2a−e, molybdenum atoms are presumably
electron rich; therefore, their reactions with electrophiles may
give products with intriguing structural features. We report, in
the present contribution, the reactions of these complexes with
electrophilic materials, including dioxygen, carbon dioxide,
carboxylic acids, and triflic acid (CF3SO3H). Furthermore, their
ability to act as a catalyst precursors for carbon dioxide fixation
has been studied in detail. The preliminary results for some of
the chemistry described herein have been previously
communicated.2b,7
Selected spectroscopic data for 3a−c are compiled in Table
2. In the 1H NMR spectra (C7D8), the hydride resonance for 3a
Table 2. Selected Spectroscopic Data for 3a−c
a
1H NMR (J, Hz)
IR ν(Mo−
31P NMR (J,
a
compd H), cm−1
δ(Mo−H)
δ(other)
Hz)
3a
3b
3c
1708
1743
1709
0.10 (tt, 1H, JPH 1.6−2.8 (m, 8H, 77 (d, 2P, JPP
= 16,40)
C2H4)
6.6−8.9 (m, 43H, 105 (d, 2P, JPP
Ph protons) = 59)
2.0−3.0 (m, 8H, 35 (br, 1P), 62
C2H4) (br, 1P)
6.2−8.0 (m, 38H, 88 (br, 1P),
Ph protons) 103 (br, 1P)
= 59)
−0.05 (tt, 1H,
JPH = 16, 39)
0.14 (tt, 1H, JPH 2.2−2.5 (m, 8H, 35−65 (br,
= 16, 39)
C2H4)
2P)
2.62 (s, 6H,
NCH3)
90−100 (br,
2P)
6.0−8.0 (m, 42H,
Ph protons)
a
Measurements at room temperature.
RESULTS AND DISCUSSION
■
2
Synthesis and Characterization of Dioxygen Com-
plexes. During our study of the chemical properties of 2a, we
found that a solution of 2a is very susceptible to air and its
color changes from yellow to green in the presence of even
trace amounts of contaminated air. This observation suggested
the formation of a dioxygen adduct of 2a. Considerable
attention has been focused on the interaction of dioxygen with
transition-metal complexes, with particular reference to the
activation of the former at the active center of a
monooxygenase such as cytochrome P450. Previously,
numerous dioxygen-coordinated transition-metal complexes
have been reported.7 Motivated by these facts, we set out to
explore the reaction between 2a−c and dioxygen.
is observed at δ 0.10 ppm (tt, JPH = 16, 40 Hz), which is at a
lower field than that for 2a by about 4 ppm. This strong
downfield shift could be interpreted as a consequence of the
extremely electrophilic character of the dioxygen ligand. The
hydride absorptions for 3b,c are found to be close to that
observed for 3a. Apparently, the substituents attached to the Si
atom have little influence on the electronic character of the
metal center. The 31P NMR spectra for 3a,c at ambient
temperature display two broad signals at low field (δ 105 ppm
for 3a and δ 95 ppm for 3c) and at high field (δ 77 ppm for 3a
and δ 50 ppm for 3c). In contrast, the spectrum for 3b reveals
four broad resonances at around 40, 60, 90, and 100 ppm.
Cooling the solution of 3b to −40 °C leads to the resolution of
these peaks into four sharp doublets that are observed at δ 102
Treating either THF or toluene solutions of the complexes
with dioxygen (1 atm) at ambient temperature produced the
2
2
2
(d, JPP = 44 Hz), 87 (d, JPP = 212 Hz), 62 (d, JPP = 44 Hz),
4942
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949

Organometallics
Article
2
and 35 ppm (d, JPP = 212 Hz) with a 1:1:1:1 peak area ratio,
indicating the presence of four inequivalent P nuclei. This
presumably occurs as a result of steric congestion at the Mo
center; the Mo−O bond lengths in 3b are significantly shorter
than those observed in 3a (vide infra).
The molecular structures of 3a,b, as determined by X-ray
crystallography are shown in Figures 1 and 2, with selected
Figure 2. Molecular structure of 3b. Phenyl rings that are not attached
to the Si atom have been omitted for clarity. Selected bond distances
(Å) and angles (deg): Mo−P1 = 2.536(2), Mo−P2 = 2.473(2), Mo−
P3 = 2.470(1), Mo−P4 = 2.514(2), Mo−Si = 2.546(2), Si−C53 =
1.915(6), Si−C28 = 1.882(6), Si−C16 = 1.898(7), O1−O2 =
1.411(7), Mo−O1 = 2.051(4), Mo−O2 = 2.095(4); P1−Mo−P2 =
79.00(6), P2−Mo−P4 = 168.35(6), P1−Mo−P3 = 169.95(6), P3−
Mo−P4 = 81.45(5), O1−Mo−O2 = 39.8(2), Mo−Si−C53 =
123.8(2), Mo−Si−C16 = 110.5(2), Mo−Si−C28 = 113.0(2), O1−
Mo−O2 = 39.8(2), P2−Mo−O1 = 75.7(1), P1−Mo−O2 = 94.1(5).
Figure 1. Molecular structure of 3a. Phenyl rings that are not attached
to the Si atom have been omitted for clarity. Selected bond distances
(Å) and angles (deg): Mo−P1 = 2.434(5), Mo−P2 = 2.495(5), Mo−
P3 = 2.422(5), Mo−P4 = 2.497(5), Mo−Si = 2.554(5), Si−C1 =
1.93(2), Si−C8 = 1.91(2), Si−C34 = 1.85(2), O1−O2 = 1.41(3),
Mo−O1 = 2.12(1), Mo−O2 = 2.11(2); P1−Mo−P2 = 80.9(2), P2−
Mo−P4 = 102.6(2), P1−Mo−P3 = 99.2(2), P3−Mo−P4 = 80.0(2),
O1−Mo−O2 = 38.9(7), Mo−Si−C1 = 124.2(5), O1−Mo−Si =
144.9(5), O1−Mo−P2 = 74.2(5), O2−Mo−P1 = 83.9(5).
substituents (C6H5 and C6F5) on the nature of the Mo−O
bond are rather weak. As a result, only the electronic effect will
be at the origin of any variations in the measured parameters,
and the amount of electron density of the Si atom will influence
directly the interaction between the metal and the dioxygen
ligand.
Particularly noteworthy is the remarkable difference in the
Mo−O bonds; the two Mo−O bond distances (mean 2.073 Å)
in 3b are significantly shorter than those in 3a (mean 2.115 Å).
This is the direct result of the electron-withdrawing C6F5 group,
which should reduce the σ-donor ability of the silyl ligand and
strengthen the Mo−O linkages. In contrast, the O−O distances
in the two compounds differ surprisingly little (1.41(3) Å for
3a; 1.411(7) Å for 3b). In general, in peroxo compounds, a σ
bond is formed by filled O(pπ)→M(dσ) bonding and back-
bonding is due to M(dπ)→O(π*).11 Taking into account these
facts, we can conclude that the presence of an electron-
withdrawing group on the silyl ligand reduces the π-back-
donation ability of the molybdenum fragment.
Treatment of the parent complex 1 or the molybdenum silyl
complex II with dioxygen in solution, only resulted in the
deterioration of the starting complex, indicating that the
structural rigidity imposed by the pentadentate framework in
3a−c is crucial in stabilizing peroxo-type dioxygen complexes of
Mo(IV).
bond distances and angles given in the captions. Both
complexes have similar structures and contain one dioxygen
ligand that is coordinated to the molybdenum atom in a side-on
fashion. The two Mo−O bond lengths in 3a (2.12(1) and
2.11(2) Å) are significantly longer than those found in common
η2-O2 complexes in which the sizes of central atoms are nearly
equal to that of Mo, e.g., [OsH(η2-O2){1,2-bis-
(dicyclohexylphosphino)ethane}2]BPh4 (2.045(8) and
2.037(8) Å)9 and [CpW(O)(η2-O2)(CH2SiMe3)] (1.68(3)
and 1.92(3) Å).10 This feature may be attributed to the strong
trans influence of the silyl ligand in 3a. The observed O−O
bond length is 1.41(3) Å; it is intermediate between those in
2−
−
peroxo (O2 = 1.49 Å) and superoxo groups (O2 = 1.33 Å)
and lies within the range reported for peroxo compounds
(1.40−1.50 Å).11
A comparison of the metrical values for 3a,b serves to
elucidate the nature of the trans influence on these complexes.
Figure 3 shows the experimental structural parameters for 3a,b.
The Mo−Si (2.546(2) Å) and Si−C(C6F5 ring) (1.915(6) Å)
distances and the Mo−Si−C(C6F5 ring) angle (123.8(2)°) in
3b are very close to the corresponding distances (2.554(5) and
1.93(2) Å) and angle (124.2(5)°) in 3a. It may be inferred
from these facts that 3a,b are sterically very similar with respect
to coordination, and therefore, the steric effects from the
Synthesis and Characterization of Monocarboxylato
Complexes. Complexes 2a−c react with 2 equiv of carboxylic
acids such as formic acid, acetic acid, and benzoic acid at room
temperature to give the monocarboxylato compounds 4a1−c3
in good yields (70−90%, Scheme 3). In Table 3, we summarize
the reaction conditions.
4943
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949

Organometallics
Article
During the reactions, the evolution of the hydrogen gas was
qualitatively monitored by GLC analysis. Selected NMR
spectroscopic data for the resulting complexes are compiled
in Table 4. In the 1H NMR spectra for 4a1−c3, hydrido signals
Table 4. Selected NMR Spectroscopic Data for 4al−c3
a
1H NMR (J, Hz)
31P NMR (J,
a
compd
δ(Mo−H)
δ(other)
Hz)
4a1
−8.01 (tt, 2H,
JPH = 17, 43)
1.9−3.0 (m, 8H, C2H4)
59 (br d, 2P,
JPP = 122)
6.1−8.9 (m, 43H, Ph protons)
88 (br d, 2P,
JPP = 122)
4a2
4a3
4b1
4b2
4b3
4c1
4c2
4c3
−7.96 (tt, 2H,
JPH = 17, 43)
1.23 (s, CH3CO), 2.0−3.0 (m,
8H, C2H4)
59 (br d, 2P,
JPP = 122)
6.2−8.9 (m, 43H, Ph protons)
88 (br d, 2P,
JPP = 122)
−7.92 (tt, 2H,
JPH = 17, 43)
1.9−3.0 (m, 8H, C2H4)
59 (br d, 2P,
JPP = 121)
6.0−8.9 (m, 48H, Ph protons)
1.8−3.2 (m, 8H, C2H4)
87 (br d, 2P,
JPP = 119)
−8.10 (tt, 2H,
JPH = 18, 43)
55 (br d, 2P,
JPP = 119)
6.0−8.0 (m, 38H, Ph protons)
87 (br d, 2P,
JPP = 119)
−8.00 (tt, 2H,
JPH = 18, 43)
1.10 (s, CH3CO), 1.8−3.2 (m,
8H, C2H4)
55 (br d, 2P,
JPP = 122)
6.3−8.0 (m, 38H, Ph protons)
87 (br d, 2P,
JPP = 122)
−8.00 (tt, 2H,
JPH = 18, 43)
2.0−3.4 (m, 8H, C2H4)
55 (br d, 2P,
JPP = 119)
6.3−8.2 (m, 43H, Ph protons)
86 (br d, 2P,
JPP = 119)
−8.10 (tt, 2H,
JPH = 17, 43)
2.2−2.6 (m, 8H, C2H4), 2.73 (s, 59 (br d, 2P,
6H, NCH3)
JPP = 121)
6.0−8.3 (m, 42H, Ph protons)
87 (br d, 2P,
JPP = 121)
Figure 3. Experimental structures for 3a,b with selected structural
parameters.
−7.90 (tt, 2H,
JPH = 17, 44)
1.70 (s, CH3CO), 2.1−2.7 (m,
8H, C2H4)
60 (br d, 2P,
JPP = 122)
2.73 (s, 6H, NCH3), 6.0−8.5 (m, 88 (br d, 2P,
42H, Ph protons) JPP = 122)
Scheme 3
−7.90 (tt, 2H,
JPH = 17, 44)
2.3−2.7 (m, 8H, C2H4), 2.75 (s, 59 (br d, 2P,
6H, NCH3)
JPP = 119)
6.0−8.3 (m, 47H, Ph protons)
88 (br d, 2P,
JPP = 119)
a
Measurement at room temperature.
appeared at around δ −8 as a triplet of triplets, corresponding
to an A2K2X spin system. The 31P{1H} NMR spectra for 4a1−
c3 appear as a typical AX system with two doublets. The
presence of only two resonances indicates that they have an
effective mirror plane containing the Mo−Si bond. The IR
spectra for complexes 4a1−c3 showed two ν(OCO) stretching
frequencies (asym and sym) near 1600 and 1330 cm−1,
respectively (Table 5). The fairly large values of Δν
(ν(OCO)asym − ν(OCO)sym), ranging from 239 to 329 cm−1,
suggest that the carboxylate ligands in these complexes
coordinate to molybdenum in a monodentate fashion, as
shown in Scheme 3.12
The basic features of the structure of these complexes
deduced from the spectra were fully confirmed by the single-
crystal X-ray diffraction analysis of 4a2. A representation of the
molecule of 4a2 is shown in Figure 4, with selected bond
distances and angles given in the caption.
Table 3. Preparation of Monocarboxylato Complexes
substrate
compd
complex (R)/amt, mmol
acid/amt, mmol
yield, %
4al
2a (Ph)/0.162
HCOOH/0.325
CH3COOH/0.536
PhCOOH/0.468
HCOOH/0.180
CH3COOH/0.210
PhCOOH/0.210
HCOOH/0.960
CH3COOH/0.960
PhCOOH/0.290
7
90
87
12
82
90
16
69
55
4a2
4a3
4bl
4b2
4b3
4cl
2a (Ph)/0.268
2a (Ph)/0.234
2b (C6F5)/0.092
2b (C6F5)/0.092
2b (C6F5)/0.092
2c (4-Me2NC6H4)/0.096
2c (4-Me2NC6H4)/0.096
2c (4-Me2NC6H4)/0.096
4c2
4c3
As expected from the aforementioned IR and NMR results,
this complex contains a pentadentate P2SiP2 ligand and a
monodentate acetate ligand. If the hydrido ligands are ignored,
4944
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949

Organometallics
Article
Table 5. Carboxyl Stretching Frequencies (cm−1) of 4al−c3
Scheme 4
a
compd
ν(CO)asym
ν(C−O)sym
Δ
4al
1618
1617
1609
1624
1639
1622
1596
1596
1696
1325
1338
1330
1311
1310
1311
1351
1357
1356
293
279
279
313
329
311
245
239
240
4a2
4a3
4bl
4b2
4b3
4cl
4c2
4c3
a
Difference between two frequencies.
2a−c appears to prevent chelate formation. However, in our
recent studies, we showed that 2a reacts with 2,4-pentanedione
to give a dihydrido complex with a chelating 2,4-pentanedio-
nate ligand, which results from the repulsion of one of the four
Mo−P bonds in the P2SiP2 ligand.15 Hence, we believe that in
2a−c the simultaneous availability of two vacant coordination
sites on the metal center is not unlikely. The acetate ligand
probably does not have a sufficiently strong π-donor character
in the present system.
Reaction of Complex 2a with Carbon Dioxide. The
formation of formato complexes 4a1, 4b1, and 4c1 led us to
study the reactivity of 2a toward carbon dioxide, since reactions
of a number of transition-metal hydrido complexes with carbon
dioxide yield formato complexes resulting from the insertion
reaction of carbon dioxide into M−H bonds.16 For example,
experiments in our laboratory allowed us to establish that the
irradiation of a benzene solution of 1 under a carbon dioxide
atmosphere yields a chelating formato-O,O′ complex (Scheme
5).17
Scheme 5
Figure 4. Molecular structure of 4a2. Phenyl rings that are not
attached to the Si atom have been omitted for clarity. Selected bond
distances (Å) and angles (deg): Mo−P1 = 2.485(2), Mo−P2 =
2.507(2), Mo−P3 = 2.496(2), Mo−P4 = 2.466(2), Mo−Si = 2.515(2),
Si−C46 = 1.911(7), Si−C47 = 1.887(7), Si−C53 = 1.869(8), Mo−O1
= 2.174(5), O1−C59 = 1.239(9), O2−C59 = 1.219(10); P1−Mo−P2
= 80.31(6), P2−Mo−P3 = 96.32(6), P3−Mo−P4 = 80.16(2), P1−
Mo−P4 = 101.77(2), Si−Mo−O1 = 143.7(1), Mo−Si−C47 =
113.2(2), Mo−Si−C53 = 122.7(2), Mo−Si−C46 = 111.0(2), Mo−
O1−C59 = 130.8(5), O2−C59−C60 = 117.4(8), O1−C59−O2 =
125.1(8).
A toluene solution of complex 2a was allowed to react under
CO2 pressure (17 atm) at room temperature, and the expected
formato complex 4a1 was isolated in 82% yield from the
reaction mixture. In addition, an NMR study provided evidence
that 4a1 loses CO2 fairly easily, regenerating 2a at 80 °C in
benzene (Scheme 6).
The insertion reaction of carbon dioxide into a Mo−H bond
under mild conditions is an outstanding feature exhibited by
complex 2a, since the parent complex 1 does not react with
carbon dioxide at ambient temperature. As described above, the
reaction between 1 and carbon dioxide requires forcing
conditions to proceed to completion. Thus, the marked
difference in their reactivities underscores the strong trans-
influencing ability of the Si fragment.
the geometry of the Mo center is a distorted octahedron. The
equatorial plane is defined by Mo1, P1, P2, P3, and P4 (∑(X−
Mo−Y) = 358.56°), and Si and O1 occupy the pseudoaxial
positions (O−Mo−Si = 143.7(1)°).
In our previous work, we showed that the parent complex 1
reacted with carboxylic acids to give cationic carboxylato
complexes in which the carboxylato is bound as a bidentate
ligand (Scheme 4).13 Furthermore, 1 reacted with allyl
carboxylate upon irradiation, giving exclusively a chelating
complex.14
Therefore, the most interesting feature of the structure is the
monodentate mode observed for the carboxylate ligands in
4a1−c3. At first sight, the restricted geometric arrangement in
Catalytic Hydrogenation of Carbon Dioxide. In view of
the straightforward insertion mode of CO2 into the Mo−H
4945
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949

Organometallics
Article
P2SiP2 girdle in complexes 2a,c,e play an important role in the
present catalytic process.
Scheme 6
A possible mechanism for the catalytic formation of DMF
should have three steps. The initial step is the hydrogenation of
CO2 to formic acid. The resulting formic acid reacts with
dimethylamine to yield ammonium formate, followed by
dehydration. Thermal condensation of dimethylamine and
formic acid is known to proceed at 100 °C;19 hence, the
dehydration is easily performed in our process.
Alternative pathways for the formation of formic acid
catalyzed by transition-metal hydrido complexes (LnM−H)
have been proposed on the basis of some experimental
evidence and theoretical calculations, as outlined in Scheme
7.20 Path A (the classical process) involves CO2 insertion into a
M−H bond, followed by a sequence of oxidative addition and
reductive elimination steps for the H2 activation process.
However, Dedieu et al. have recently suggested a different
pathway (σ-bond metathesis) in which no change in the
oxidation state of the metal center is required and the
formation of formic acid is a result of a [2 + 2] addition of
H2 on a metal formato intermediate obtained after CO2
insertion (path B of Scheme 7).21 The present catalytic process
can be explained by either path A or path B.
bond of 2a, we decided to investigate the capability of 2a to
catalyze the hydrogenation of CO2 itself. Many studies have
previously shown that metal formato complexes are assumed to
be the active catalytic intermediates in this process.18
The catalytic reactions were performed at 110 °C in a
stainless-steel autoclave under CO2/H2 gas mixtures in the
presence of dimethylamine to provide N,N-dimethylformamide
(DMF). The results of the hydrogenation of CO2 using 2a−c,e
are summarized in Table 6. Additionally, in order to compare
Table 6. Effect of Catalyst on TON of Hydrogenation of
a
CO3
Dedieu et al. established that the reductive elimination of
formic acid from an intermediate formato hydrido complex X is
the rate-limiting step for path A, whereas the liberation of
formic acid from a formic acid complex Y is the rate-limiting
step for path B.21 Our study has shown that the thermolysis of
formato hydrido complex 4a1 leads only to the evolution of
CO2 and regeneration of 2a; therefore, complex 4a1 does not
reductively eliminate formic acid (Scheme 6). Thus, we believe
that a σ-bond metathesis reaction occurs in the present process.
Although we must await further investigations to elucidate the
correct mechanism of our process, we can consider the present
experimental results useful because the trans effect of the silyl
ligand can efficiently operate in both these pathways. As
discussed in the previous section, the studies performed on the
dioxygen complexes have shown that the perfluoro phenyl
group in 3b decreases substantially the trans influence
properties of the Si ligand and causes a dramatic strengthening
of the Mo−O linkages. Accordingly, it seems likely that both
the reductive elimination of formic acid and the liberation of
formic acid do not occur easily in intermediate complexes
derived from 2b. In direct contrast to 2b, the incorporation of a
Me2N group at the 4-position of Ph is expected to elicit the
opposite effect; hence, this leads to 2c being the most efficient
catalyst precursor.
Reactions of 2a,d with Triflic Acid. The reactions of 2a−
c with relatively weak acids, i.e., carboxylic acids, were observed
to result in the formation of monocarboxylato molybdenum
complexes. Subsequently, we focused on the further reaction
chemistry of them with strongly acidic CF3SO3H (triflic acid,
TfOH).
Reactions of metal alkyl and hydrido complexes with strong
protonic acids, such as triflic acid, have provided a general
synthetic route to metal complexes with weakly coordinated
ligands.22 Furthermore, in the case of 2a, there is a possibility of
breakage of the silyl ligand, since the reaction of PhSiH3 with
triflic acid results in the preferential cleavage of the Si−Ph
bond, generating H3SiOTf and benzene.23 This process has
proved to be a very simple, energetically accessible reaction.
Consequently, it seems to be difficult to make reliable
b
entry no.
cat. (R)
2a (Ph)
TON
1
2
3
4
5
6
7
92
1
2b (C6F5)
2c (4-Me2NC6H4)
2e (n-C6H13)
1
130
110
0
c
I
0
c
II
9
a
Conditions: 110 °C, 20−25 atm of CO2, 30−40 atm of H2, 50 mmol
of HN(CH3)2 (25 mL, 2.0 M solution in toluene), 0.2 mmol of
catalyst. TON = turnover number = mol of product (DMF)/mol of
b
c
catalyst. See Chart 1.
the effects of the Si ligand system on the catalytic reactivity,
parent complex 1, germanium congener I, and complex II
containing a tridentate ligand were included in this study. The
results are summarized in Table 6.
As can be seen from the table, there is a wide range of values
obtained for the catalytic activity. Complex 2c is the most
active; the use of 2c allows the reaction to reach a turnover
number (TON) of 130. The presence of electron-donating
substituents on the Si ligand in this catalyst system may lead to
an increase in catalytic activity, because complexes 2c,e are
more reactive than complex 2a. Furthermore, this increase
seems to parallel the increase in electron density of the Si atom.
Consistent with this premise, no catalytic reaction was found
with 2b. The electron-withdrawing substituent dramatically
suppresses the activity. Interestingly, complex II is much less
active than complex 2a. Moreover, as can be appreciated from
the table, both complexes 1 and I are virtually inactive. These
results strongly suggest that both the Mo−Si linkage and the
4946
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949

Organometallics
Article
Scheme 7
predictions for the final products of the reaction between 2a
and triflic acid.
Table 7. Selected Structural Data for 5a,d
a
IR/cm−1
1H NMR (J, Hz)
The treatment of a brown THF solution of complex 2a with
1.2 equiv of triflic acid at −78 °C resulted in an obvious color
change to red, but the resulting products were not sufficiently
stable under these conditions and attempts to isolate the
products were unsuccessful. Thus, the desired products were
converted into the more stable hydroxide complex 5a on
treatment with aqueous NaOH (Scheme 8). The cyclohexyl
analogue 5d can be prepared in a similar manner.
ν(Mo− ν(Mo−
31P NMR (J,
a
compd
H)
OH)
3452
δ(Mo−H)
δ(other)
Hz)
5a
1810
−7.25 (tt, 2H,
JHP = 16, 46)
2.0−2.8 (m,
8H, C2H4)
54 (d, 2P,
JPP = 144)
88 (d, 2P,
JPP = 144)
5d
1825
3484
−7.59 (tt, 2H,
JHP = 16, 46)
1.6−2.8 (m,
8H, C2H4)
51 (d, 2P,
JPP = 141)
89 (d, 2P,
JPP = 141)
Scheme 8
a
Measurement at room temperature.
Crystals of 5d were grown from solution in a benzene/
hexane solvent mixture at 0 °C, and the molecular structure of
5d was confirmed by a single-crystal X-ray diffraction study. A
molecular structure of 5d is shown in Figure 5, with selected
bond distances and angles given in the caption. We note that
the accuracy of these metrical parameters is not completely
Selected spectroscopic data for 5a,d are compiled in Table 7.
IR absorptions near 1820 and 3460 cm−1 are respectively
1
attributable to ν(Mo−H) and ν(Mo−OH). In the H NMR
spectra, hydrido signals appeared at around δ −7.5 as a triplet
of triplets corresponding to an A2K2X spin system. The
31P{1H} NMR spectra for these complexes appear as a typical
AX system with two doublets. The presence of only two
resonances indicates that the molecules have an effective mirror
plane containing a Mo−Si bond. The features of the NMR
spectra for 5a,d are very similar to those for the above
monocarboxylato complexes, suggesting that these complexes
have analogous structures.
Figure 5. Molecular structure of 5d. Phenyl rings that are not attached
to the Si atom have been omitted for clarity. Selected bond distances
(Å) and angles (deg): Mo−P1 = 2.462(3), Mo−P2 = 2.486(5), Mo−
P3 = 2.475(4), Mo−P4 = 2.510(4), Mo−Si = 2.519(4), Si−C7 =
1.922(17), Si−C1 = 1.884(13), Si−C33 = 1.903(16), Mo−O1 =
2.258(9); P1−Mo−P2 = 81.31(15), P1−Mo−P3 = 98.84(15), P3−
Mo−P4 = 80.26(14), P2−Mo−P4 = 97.29(15), Si−Mo−O1 =
147.44(2), Mo−Si−C7 = 110.81(5), Mo−Si−C1 = 125.79(5), and
Mo−Si−C33 = 109.23(4).
4947
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949

Organometallics
Article
satisfactory. The crystal of 5d was not good enough for
refinement (R = 0.1330), and this problem prevented a high-
quality structure determination. However, the substituents are
rather well-defined and have their expected geometries.
If the hydrido ligands are ignored, the coordination geometry
around the Mo center is a distorted octahedron. The equatorial
plane is defined by Mo1, P1, P2, P3, and P4; Si and O occupy
the pseudoaxial positions. The Mo−O separation of 2.258(9) Å
in 5d is much longer than the typical value for an MO bond
in oxo compounds (∼1.59−1.66 Å).24 Furthermore, this Mo−
O bond length is comparable with the value of 2.174(5) Å in
4a2. These observations strongly suggest a single bond between
molybdenum and oxygen. Thus, the oxygen atom, O1, is
actually a hydroxide, OH. Indeed, the IR spectra for 5a,d were
consistent with the presence of a characteristic O−H stretching
frequency between 3452 (5a) and 3484 cm−1 (5d) (vide
supra).
As mentioned previously, the reactions between triflic acid
and complex 2a may proceed by two pathways to give the main
products including (a) protonation at the molybdenum atom
and (b) cleavage of the Si−Ph bond. Our results have shown
that the end product corresponds structurally to the hydroxide
coordination adduct, and hence, the initial stage of this reaction
would be protonation at the metal center to give a triflato
complex.
dimethylamine and 2a. In contrast, when the experiment is
performed under identical conditions in the presence of 1 as a
catalyst there is no measurable formation of DMF. To our
knowledge, complex 2a is the first example of a molybdenum
phosphine compound that has proved to be a considerably
efficient catalyst for a carbon dioxide fixation reaction. In
summary, the introduction of the silyl group into the dppe
ligand system dramatically changes the reactivity of the
molybdenum polyhydrido complexes and remarkably enhances
their catalytic performance. Additionally, we investigated the
influence of different substituents, directly attached to the Si
atom of the silyl ligand, on their chemical behavior. We have
revealed that electron-donating substituents increase the trans-
influencing ability of the Si group and lead to an increase in the
catalytic activity.
Owing to their important role in several metal-catalyzed
processes, including olefin hydrosilylation, silane alcoholysis,
and dehydropolymerization of silanes, transition-metal silyl
complexes have received a great deal of attention. The present
investigations of the reactivities of 2a−e suggest that transition-
metal silyl complexes deserve considerable interest with respect
to other catalytic processes. The new complexes reported here
provide us with the basis for further development of the
chemistry of transition-metal silicon complexes in the areas of
small-molecule activation and catalysis.
We note here that a similar reaction of 2a with trifluoroacetic
acid, followed by treatment with aqueous NaOH, led
exclusively to an η1-O2CCF3 complex, while complex 5a was
not able to be detected.25 Perhaps this is not surprising, given
that the triflato complex possesses a greatly enhanced reactivity
in comparison with the η1-O2CCF3 complex. Thus, the triflato
precursor is an attractive candidate for preparing novel types of
complexes bearing polydentate phosphinoalkyl−silyl ligands.
Transition-metal complexes containing a triflate ligand have
been studied extensively because they may be versatile and
important reagents in mechanistic studies and organometallic
synthesis. Efforts in this direction are continuing.
In addition, hydroxide complexes are interesting in their own
right, since they are central to catalytic reactions such as the
Wacker process and the water-gas shift reaction. Consequently,
further studies investigating the catalytic performance of these
complexes are in progress.
EXPERIMENTAL SECTION
■
General Procedures. All manipulations were performed using
standard Schlenk techniques under purified argon. All reagents were
purchased from commercial suppliers and used as received without
further purification. All solvents were dried by standard methods and
were stored under argon. Complexes 2a−e were synthesized according
to procedures previously reported.2 NMR data were recorded on
either a JEOL-JNM-EX270 or a JEOL-JNM-AL400 instrument.
31P{1H} NMR peak positions were referenced to external PPh3.
Synthesis and Characterization of Dioxygen Complexes. A
pale yellow solution of 2a (217 mg, 0.217 mmol) in dry THF (22 mL)
turned green on exposure to 1 atm of dioxygen with stirring for 2 h at
ambient temperature. After evaporation of the solvent under vacuum,
the residue was washed with hexane and then recrystallized from
THF/hexane to afford 3a (213 mg, 87%) of greenish crystals. Anal.
Calcd for C66H68MoO4P4Si (3a·2THF): C, 67.57; H, 5.84. Found: C,
67.24; H, 5.16. This procedure is also applicable to the synthesis of the
other dioxygen complexes (3b,c). Experiments given in Table 1 were
carried out under essentially the same conditions. Spectral data for
3a−c are summarized in Table 2. Satisfactory microanalytical data for
3b,c were not obtained after several attempts.
Synthesis and Characterization of Monocarboxylato Com-
plexes. To a stirred solution of 2b (100 mg, 0.092 mmol) in dry
toluene (10 mL) was added glacial acetic acid (11.8 μL, 0.210 mmol)
drop by drop with a syringe at ambient temperature. After further
stirring for 10 h, the color of the mixture changed from yellow to
orange. Addition of excess hexane to the resulting mixture gave a
yellow precipitate. After removal of the supernatant, the precipitate
was dried under reduced pressure and then washed with hexane. The
resulting solid was dried in vacuo to give 4b2 (85 mg, 82%). This
procedure is also applicable to the synthesis of the other
monocarboxylato complexes. Experiments given in Table 3 were
carried out under essentially the same conditions. Spectral data for
4a1−c3 are summarized in Table 4. Brown crystals of 4a2, suitable for
an X-ray structure determination, were grown from THF/hexane
solution at −10 °C. Complexes 4a1−c3 were too unstable to allow
satisfactory elemental analyses. Repeated purifications by recrystalliza-
tion or reprecipitating led to partial decomposition with concomitant
formation of small amounts of impurities.
CONCLUSIONS
■
Several important conclusions can be made from our results.
Complexes 2a−e are found to exhibit significant differences in
their reactions with electrophilic materials in comparison to the
parent complex [MoH4(dppe)2] (1). These differences are
interpreted in terms of a combination of the strong trans-
labilizing ability of the Si group and the unusual quintuply
chelated structure of these complexes. Unlike 1, which
decomposes in the presence of dioxygen, stable η2-O2
complexes 3a−c are formed by the reactions of 2a−c with
dioxygen. The treatment of 2a−c with carboxylic acids gives
neutral molybdenum complexes 4a1−c3 with a monodentate
carboxylate ligand, while the analogous reaction with 1 leads to
a cationic carboxylato complex possessing a chelating bidentate
structure. Carbon dioxide reacts reversibly with 2a at room
temperature to form formato complex 4a1, which is also
obtained by the reaction of 2a with formic acid. In the case of 1,
this similar reaction does not occur. Furthermore, we have
found that DMF is prepared by the catalytic hydrogenation of
carbon dioxide under pressure at 110 °C in the presence of
Reaction of Complex 2a with Carbon Dioxide. 2a (310 mg,
0.310 mmol) was dissolved in dry THF (35 mL) in a Schlenk flask
4948
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949

Organometallics
Article
under an argon atmosphere. The solution was transferred into an
autoclave equipped with a stir bar. The autoclave was then pressurized
to 19 atm with CO2 and was stirred at ambient temperature for 3 h.
The reactor was vented at ambient temperature, and the reaction
mixture was transferred into a Schlenk flask under an argon
atmosphere. Hexane (15 mL) was then added, and the mixture was
allowed to stand for 12 h at 0 °C, whereupon a yellow precipitate
separated out. The product was collected by filtration, washed with
hexane (10 mL × 2), and then dried in vacuo to give a yellow solid,
which was identified as 4a1 (265 mg, 82%) on the basis of its
spectroscopic properties.
Catalytic Hydrogenation of Carbon Dioxide. In a typical
experiment, a catalyst (0.2 mmol) and dimethylamine (25 mL, 50
mmol, 2.0 M solution in toluene) were combined in a Schlenk flask
under an argon atmosphere. Then, the solution was transferred into an
autoclave equipped with a stir bar. The autoclave was sealed,
pressurized to 50 atm with CO2/H2 (3/2), and heated to 110 °C.
After 24 h of stirring, the autoclave was vented and the reaction
mixture was transferred into a Schlenk flask. The volatiles were
vacuum-transferred into another vessel and then analyzed by GLC.
Yields of DMF obtained were determined relative to an internal
mesitylene standard. The results are collected in Table 6.
Reaction of 2a,d with Triflic Acid. To a stirred solution of 2d
(280 mg, 0.280 mmol) in dry THF (32 mL) was added neat triflic acid
(0.030 mL, 0.34 mmol) drop by drop with a syringe at −78 °C. The
solution was warmed to ambient temperature and stirred for 1 h,
during which time a gradual color change from brown to red was
observed. Subsequently, the solution was quenched by the addition of
aqueous NaOH solution (1.5 equiv). The resulting yellow solution was
filtered, and all volatiles were removed in vacuo. The residue was
washed with hexane and then dried in vacuo to give 5d (240 mg,
86%). This procedure is also applicable to the synthesis of 5a. Spectral
data for 5a,d are summarized in Table 7. Prismatic crystals of 5d,
suitable for an X-ray structure determination, were grown from
benzene/hexane solution at −10 °C. The instability of 5a,d has
prevented microanalysis.
24. (b) Tilley, T. D. In The Silicon-Heteroatom Bond; Patai, S.,
Rappaport, Z., Eds.; Wiley: New York, 1991; Chapter 10. (c) Corey, J.
Y.; Braddock-Wilking, J. Chem. Rev. 1999, 99, 175.
(2) (a) Zhou, D.-Y.; Minato, M.; Ito, T.; Yamasaki, M. Chem. Lett.
1997, 1017. (b) Zhou, D.-Y.; Zhang, L.-B.; Minato, M.; Ito, T.;
Osakada, K. Chem. Lett. 1998, 187. (c) Kubas, G. J. In Metal
Dihydrogen and σ-Bond Complexes; Kluwer: New York, 2001; pp 339−
340. (d) Minato, M.; Zhou, D.-Y.; Zhang, L.-B.; Hirabayashi, R.;
Kakeya, M.; Matsumoto, T.; Harakawa, A.; Kikutsuji, G.; Ito, T.
Organometallics 2005, 24, 3434.
(3) (a) Auburn, M. J.; Holmes-Smith, R. D.; Stobart, S. R. J. Am.
Chem. Soc. 1984, 106, 1314. (b) Auburn, M. J.; Grundy, S. L.; Stobart,
S. R.; Zaworotko, M. J. J. Am. Chem. Soc. 1985, 107, 266. (c) Auburn,
M. J.; Stobart, S. R. Inorg. Chem. 1985, 24, 318. (d) Holmes-Smith, R.
D.; Stobart, S. R.; Vefghi, R.; Zaworotko, M. J.; Jochem, K.; Cameron,
T. S. J. Chem. Soc., Dalton Trans. 1987, 969. (e) Okazaki, M.;
Ohshitanai, S.; Tobita, H.; Ogino, H. Chem. Lett. 2001, 952.
(f) Okazaki, M.; Tobita, H.; Ogino, H. J. Chem. Soc., Dalton Trans.
2002, 2061. (g) Sangtrirutnugul, P.; Stradiotto, M.; Tilley, T. D.
Organometallics 2006, 25, 1607.
(4) (a) Takaya, J.; Iwasawa, N. J. Am. Chem. Soc. 2008, 130, 15254.
(b) MacInnis, M.; McDonald, R.; Ferguson, M. J.; Tobisch, S.;
Turculet, L. J. Am. Chem. Soc. 2011, 133, 13622. (c) Lee, Y.; Peters, J.
J. Am. Chem. Soc. 2011, 133, 4438.
(5) (a) Chatt, J.; Eaborn, C.; Ibekwe, S. J. Chem. Soc., Chem. Commun.
1966, 700. (b) McWeeny, R.; Mason, R.; Towl, A. D. C. Discuss.
Faraday Soc. 1969, 47, 20. (c) Haszeldine, R. N.; Parish, R. V.;
Setchfield, J. H. J. Organomet. Chem. 1973, 57, 279.
(6) (a) Minato, M.; Nishiuchi, J.; Kakeya, M.; Matsumoto, T.;
Yamaguchi, Y.; Ito, T. Dalton Trans. 2003, 483. (b) Minato, M.;
Kikutsuji, G.; Kakeya, M.; Osakada, K.; Yamasaki, M. Dalton Trans.
2009, 7684.
(7) Minato, M.; Zhou, D.-Y.; Sumiura, K.; Hirabayashi, R.;
Yamaguchi, Y.; Ito, T. Chem. Commun. 2001, 2654.
(8) (a) Dickman, M. H.; Pope, M. T. Chem. Rev. 1994, 94, 569.
(b) Kitajima, N.; Fujisawa, K.; Fujimoto, C.; Moro-oka, Y.; Hashimoto,
S.; Kitagawa, T.; Toriumi, T.; Tatsumi, K.; Nakamura, A. J. Am. Chem.
Soc. 1992, 114, 1277.
(9) Mezzetti, A.; Zangrando, E.; Del Zotto, A.; Rigo, P. Chem.
Commun. 1994, 1597.
ASSOCIATED CONTENT
* Supporting Information
■
S
Tables, text, figures, and CIF files giving crystal data and
experimental parameters for compounds 3a,3b, 4a2, and 5d,
NMR (1H, 13C, and 29Si) and IR data for all compounds, and
NMR (1H, 13C, 31P, and 29Si) spectra for 3b,c and 4a1−4c3
(S3). This material is available free of charge via the Internet at
http://pubs.acs.org. The data for 3a, 3b, 4a2, and 5d have also
been deposited with Cambridge Crystallographic Data center,
as CCDC Nos. 833579 (3a), 822146 (3b), 172503 (4a2), and
81833 (5d).
(10) Legzdins, P.; Phillips, E. C.; Rettig, C. J.; San
́
chez, L.; Trotter, J.;
Yee, V. C. Organometallics 1988, 7, 1877.
(11) Cotton, F. A.; Wilkinson, G. In Advanced Inorganic Chemistry,
5th ed.; Wiley: New York, 1988; p 469.
(12) Nakamoto, K. In Infrared and Raman Spectra of Inorganic and
Coordination Compounds, 3rd ed.; Wiley: New York, 1977; pp 230−
233.
(13) Ito, T.; Takahashi, A.; Tamura, S. Bull. Chem. Soc. Jpn. 1986, 59,
3489.
(14) Ito, T.; Matsubara, T.; Yamashita, Y. J. Chem. Soc., Dalton Trans.
1990, 2407.
(15) Ito, T.; Shimada, K.; Ono, T.; Minato, M.; Yamaguchi, Y. J.
AUTHOR INFORMATION
Corresponding Author
*E-mail: minato@ynu.ac.jp. Tel: +81-45-339-3933. Fax: +81-
45-339-3933.
■
Oranomet. Chem. 2000, 611, 308.
(16) Sneeden, R. P. A. In Comprehensive Organometallic Chemistry;
Wilkinson, G., Ed.; Pergamon Press: Oxford, U.K., 1982; Vol. 8, pp
225−283.
(17) Ito, T.; Matsubara, T. J. Chem. Soc., Dalton Trans. 1988, 2241.
(18) Jessop, P. G.; Ikariya, T.; Noyori, R. Chem. Rev. 1995, 95, 259
and references cited therein.
(19) Mitchell, J. A.; Reid, E. E. J. Am. Chem. Soc. 1931, 53, 1879.
(20) Jessop, P. G.; Hsiao, Y.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc.
1994, 116, 8851.
(21) Hutschka, H.; Dedieu, A.; Eichberger, M.; Fornika, R.; Leitner,
W. J. Am. Chem. Soc. 1997, 119, 4432.
Present Address
§The Institute of Scientific and Industrial Research, Osaka
University, Mihogaoka, Ibaraki, Osaka 567-0047, Japan.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are grateful to Dr. M. Yamasaki of Rigaku Corp. for X-ray
structure analyses.
■
(22) Beck, W. Inorg. Synth. 1990, 28, 1 and references cited therein.
(23) Molander, G. A.; Corrette, C. P. Organometallics 1998, 17, 5504.
(24) Cotton, F. A.; Wilkinson, G. In Advanced Inorganic Chemistry,
5th ed.; Wiley: New York, 1988; p 464.
REFERENCES
■
(1) (a) Tilley, T. D. In The Chemistry of Organic Silicon Compounds;
Patai, S., Rappaport, Z., Eds.; Wiley: New York, 1989; Part 2, Chapter
(25) Mine, S.; Minato, M. To be submitted for publication.
4949
dx.doi.org/10.1021/om200646c | Organometallics 2012, 31, 4941−4949