Reactions of molybdenum hydrides with organochlorosilanes: Silicon-silicon bond formation under mild conditions

Article abstract of DOI:10.1246/cl.140203

Reactions of molybdenum hydrides containing polydentate phosphinoalkylsilyl ligands with a number of chlorosilanes have been investigated; this has led to the discovery of a novel type of a dechlorinative Si-Si coupling reaction.

Full text of DOI:10.1246/cl.140203

Received: March 10, 2014 | Accepted: April 3, 2014 | Web Released: April 10, 2014
CL-140203
Reactions of Molybdenum Hydrides with Organochlorosilanes:
Silicon-Silicon Bond Formation under Mild Conditions
Takahiro Asaeda, Joo Yeon Lee, Kyosuke Watanabe, and Makoto Minato*
Department of Materials Chemistry, Graduate School of Engineering, Yokohama National University,
79-5 Tokiwadai, Hodogaya-ku, Yokohama, Kanagawa 240-8501
(E-mail: minato@ynu.ac.jp)
Reactions of molybdenum hydrides containing polydentate
Ph
phosphinoalkylsilyl ligands with a number of chlorosilanes have
been investigated; this has led to the discovery of a novel type of
a dechlorinative Si-Si coupling reaction.
Ph
Si
H
Ph
P
P
PhSiH3
H2C
CH2
Mo
H2C
Ph
CH2
P
110 °C −H2
P
Ph
Ph
H
H
Ph
Ph
Ph
Ph
H
H
H
Ph
P
P
2
H2C
CH2
Mo
Silicon-silicon bond-forming reactions are of fundamental
importance in organosilicon chemistry.1 Interest continues to
be directed toward the development of efficient reactions that
provide a practical procedure for the synthesis of polysilanes.2
The Würtz-type reaction, which involves the dechlorinative
coupling of a chlorosilane using Na, represents a convenient
route to high-molecular-weight polysilanes.3 However, the
synthetic usefulness of this procedure is somewhat limited
because it cannot tolerate a number of functional groups that
are sensitive to reduction. Recently, some alternatives to this
commonly used reaction have been reported. Examples of such
systems include Mg/ZnCl2,4 Sm/SmI2,5 low-valence titanium,6
pyrophoric lead,7 and yttrium metal.7 To our knowledge,
however, homogeneously promoted reactions using metal
hydrides that are comparable to these heterogeneous systems
have not been developed. Here, we report the ability of
chlorosilanes to undergo reductive coupling to afford the
corresponding disilanes using molybdenum hydrides.
H2C
Ph
CH2
Ph
P
P
Ph
Ph
H
Ph
Ph
Ph
Ph
CH2
Ph
Si
H
P
P
1
H2C
Ph2SiH2
Mo
H2C
Ph
CH2
Ph
110 °C −H2
P
P
H
H
Ph
Ph
3
Scheme 1. Reaction of 1 with organosilanes.
r. t.
110 °C
1 + PhSiH2Cl
Ph
Si
Cl
Ph
Ph
Ph
Ph
Ph
Cl
Cl
H
Ph
CH2
CH2
Ph
Ph
P
P
P
P
H2C
H2C
CH2
CH2
Ph
Mo
Mo
+
PhSiH3
H2C
Ph
H2C
Ph
P
P
P
P
H
H
Cl
H
Ph
Ph
Ph
Ph
not observed
5
4
Hydride complex 2 was inadvertently produced during the
course of our studies on the thermal reaction of [MoH4(dppe)2]
(1, dppe: Ph2PCH2CH2PPh2) with PhSiH3.8 This novel trans-
formation, which involves Si-H bond cleavage with concom-
itant selective activation of the ortho C-H bonds of phenyl
groups in the dppe ligands, was shown to proceed via a silylene
metal intermediate. The result prompted us to conduct an
extensive study of the reactions between 1 and a variety of
organosilanes. For example, with Ph2SiH2, complex 3 contain-
ing a tridentate ligand was additionally formed (Scheme 1).9
Deviating behavior was observed for PhSiH2Cl, which was
found to react with 1 at 110 °C in a manner similar to that
described above, resulting in the formation of an analogous
chloride 4.10 It is interesting that, unlike PhSiH3, the reaction of
PhSiH2Cl with 1 proceeded at room temperature, yielding the
known complex, cis-[MoH2Cl2(dppe)2] (5, Scheme 2).11
This reaction might be expected to afford PhSiH3 as a by-
product as simple metathesis of the chloride in PhSiH2Cl with
hydride leads to its formation. Detailed 1H NMR analysis of
the crude products revealed several peaks in the Si-H region,
indicating that a complex mixture of hydroorganosilanes was
obtained. However, no significant amount of PhSiH3 was
detected.12 This unexpected finding prompted us to explore the
reactions between 1 and chlorosilanes at room temperature.
In order to unequivocally determine the structure of the
organosilyl species resulting from this transformation, we
Scheme 2. Reaction of 1 with PhSiH2Cl.
conducted the reaction of 1 with the simpler chlorosilane,
Me3SiCl, at room temperature. Within 1 h, complete conversion
to 5 and Me3SiSiMe3 was found to have occurred, as verified
using H and 29Si NMR spectra (eq 1).13
1
r. t.
ð1Þ
1 + 4 Me3SiCl
2 Me3SiSiMe3
+ 5
The yield of Me3SiSiMe3, as determined via 1H NMR
integration with 1,4-dioxane as the internal standard, was found
to be >180% (relative to 1); thus, complex 1 reacted with
4 equiv of Me3SiCl to afford 2 equiv of the product.
It is interesting to speculate on the reason for the preferential
formation of the Si-Si bond instead of the reduction of the Si-Cl
bond in Me3SiCl. The hydridic property of a transition-metal
hydride depends on the charge, nature of the spectator ligands,
and in particular, location of the metal in the periodic table.14
The respective Pauling electronegativity values for H and
Mo(IV) are 2.20 and 2.24; hence, the facile transfer of H from
1 to Me3SiCl is unlikely to take place.
While the mechanistic details of the present reaction are not
completely clear, it likely occurs through a series of intermedi-
ates that contain a metal-silicon σ bond. The robustness of
a transition metal-silicon bond is well documented.15 The
¹
Chem. Lett. 2014, 43, 1005–1007 | doi:10.1246/cl.140203
© 2014 The Chemical Society of Japan | 1005

Table 2. Formation of disilanes using 2 or 3
2 or 3
2 R3SiCl
R3SiSiR3
Entry
R3SiCl
Complex
¸
1/2/h
Yield/%a
1
2
3
4
PhMe2SiCl
2
3
3
3
0.4
0.3
®
92b
96b
48b
72c
PhMe2SiCl
Ph2MeSiCl
Ph3SiCl
®
aIsolated yield. bThe resulting disilanes were identified by
comparison of their 1H and 29Si NMR spectra with those
reported in the literatures for these compounds (see refs 5 and
Scheme 3. Mechanistic proposal.
c
20). The IR, 1H, and 29Si NMR spectra are consistent with
those of a commercial sample (Tokyo Chemical Industry).
Table 1. Dechlorinative coupling of chlorosilanes
r. t.
1
+
4 R3SiCl
R3SiCl
2 R3SiSiR3
Time/h
+
5
Entry
Yield/%
1
2
3
4
Me3SiCl
1
24
72
72
99a
PhMe2SiCl
Ph2MeSiCl
Ph3SiCl
75b
NRc
NRc
1
b
c
aDetermined by H NMR. Isolated yield. NR: no discernible
Chart 1. Possible structures of 6 and 7.
reaction.
oxidative addition of silyl chlorides to unsaturated transition
metal is a typical route for the formation of metal-silicon
bonds.16 However, it is unlikely that such an active intermediate
is involved in the present process, as the room temperature
reaction of 1 cannot induce the liberation of H2; therefore, the
formation of the 16e intermediate, [MoH2(dppe)2], would not
occur under such conditions. Accordingly, an alternative
mechanism in which the Si-Cl bond in Me3SiCl can be
activated without requiring a prelude to H2 loss from 1 is
conceivable.
the reactivity of 1 decreased as the alkyl groups became more
bulky. This observation is in accordance with the mechanism
shown in Scheme 3, as the rate of σ-bond metathesis should be
sensitive to the size of the silane.17
Next, we proceeded by examining the reactions of the above
complexes 2 and 3 with bulky chlorosilanes because they are
indeed more reactive than 1 toward a number of substrates
because of the strong trans-influencing ability of the silicon-
containing fragment.19 For comparison, the rates of the reaction
of PhMe2SiCl with 2 and 3 were measured; the half-life (¸1/2
)
We assume that the reaction occurs by a σ-bond metathesis
mechanism, as has been described for the dehydrocoupling of
organosilanes catalyzed by group 4 metallocenes.17 A likely
mechanism for the reaction is outlined in Scheme 3. This
pathway goes through a four-center transition state where the
oxidation state of the metal does not change. Complex 1 reacts
with 1 equiv of Me3SiCl by the elimination of HCl to yield an
of the silane in each run is also presented in Table 2 (Entries 1
and 2).
As expected, reactions of both 2 and 3 with bulky
chlorosilanes proceeded readily to form the corresponding
disilanes, with 3 reacting 31 times faster than 1. The products
were obtained in moderate yields (48-96%, based on 2 or 3).
The relative reactivity was found to be 3 > 2, because the
reaction rate using 3 was faster than that using 2.
1
intermediate A. H NMR spectral monitoring of the experimen-
tal run containing 1 and Me3SiCl did not detect any presence
of A, suggesting that it rapidly reacted with another Me3SiCl
upon its formation. Subsequent σ-bond metathesis reaction of
A with 1 equiv of Me3SiCl proceeds via the formation of
Me3SiSiMe3 and monochloride B, which further reacts with
2 equiv of Me3SiCl by similar σ-bond metathesis reactions to
yield Me3SiSiMe3 and 5.
In order to explore the utility of this dechlorinative
coupling, we decided to examine the reactivity of 1 toward
several other organochlorosilanes (Table 1).18
A puzzling feature of the present system is the observation
that unlike 1, no further reaction ensued when 2 equiv of
chlorosilane were consumed. Namely, the reaction stoichiometry
indicates that both 2 and 3 react with 2 equiv of chlorosilane
to afford 1 equiv of the product. It is possible that the steric
constraints in the complexes are so great that the reaction of
a third mole of chlorosilane is unfavorable; the P-Si ligands
would prevent the approach of the incoming chlorosilane to the
vicinity of the metal in 2 and 3 from the same side of the silicon
atom.
As indicated above, treatment of 1 with Me3SiCl at room
temperature resulted in the rapid formation of Me3SiSiMe3.
On the other hand, reaction of 1 with PhMe2SiCl proceeded
sluggishly to completion; the half-life of the silane was 9.2 h.
Further, Ph2MeSiCl and Ph3SiCl were found to be highly
unreactive with 1 under the reaction conditions. It is clear that
By analogy of the above reactions between 1 and chloro-
silanes, it is conceivable that monochloride complex 6 or 7 was
concurrently formed as a by-product (Chart 1). The 1H NMR
spectra of the resulting reaction mixtures indeed showed
completely new signals in the hydride region, suggesting the
formation of novel hydrido chloride complexes.
1006 | Chem. Lett. 2014, 43, 1005–1007 | doi:10.1246/cl.140203
© 2014 The Chemical Society of Japan

Table 3. Selected spectroscopic NMR data for 6, I, and II
M. Ishikawa, Macromolecules 2004, 37, 367. c) S. J. Holder,
M. Achilleos, R. G. Jones, Macromolecules 2005, 38, 1633.
S. Kashimura, Y. Tane, M. Ishifune, Y. Murai, S. Hashimoto,
T. Nakai, R. Hirose, H. Murase, Tetrahedron Lett. 2008, 49,
269.
a) I. Kamiya, K. Iida, N. Harato, Z.-f. Li, Y. Tomisaka, A.
Ogawa, J. Alloys Compd. 2006, 408-412, 437. b) Z. Li, K.
Iida, Y. Tomisaka, A. Yoshimura, T. Hirao, A. Nomoto, A.
Ogawa, Organometallics 2007, 26, 1212.
Ph
Ph
4
5
H Si
Ph
P
P
H2C
H2C
Ph
CH2
Mo
CH2
Ph
P
P
H
X
Ph
Ph
X = Cl (6), OAc (I), OH (II)
Compd 1H ¤ (Mo-H)
31P (J, Hz)
6
7
8
G. Lai, Z. Li, J. Huang, J. Jiang, H. Qui, Y. Shen,
J. Organomet. Chem. 2007, 692, 3559.
R. E. Benfield, R. H. Cragg, R. G. Jones, A. C. Swain,
J. Chem. Soc., Chem. Commun. 1992, 1022.
M. Minato, D.-Y. Zhou, L.-B. Zhang, R. Hirabayashi, M.
Kakeya, T. Matsumoto, A. Harakawa, G. Kikutsuji, T. Ito,
Organometallics 2005, 24, 3434.
D.-Y. Zhou, L.-B. Zhang, M. Minato, T. Ito, K. Osakada,
Chem. Lett. 1998, 187.
6
I
II
¹8.06 (tt, JPH = 17, 43 Hz) 56 (d, 122), 88 (d, 122)
¹7.96 (tt, JPH = 17, 43 Hz) 59 (d, 122), 88 (d, 122)
¹7.25 (tt, JPH = 16, 46 Hz) 54 (d, 144), 88 (d, 144)
The tentative structure of 6 was estimated by spectral
comparison with two structurally related compounds I and II
that had been fully characterized by spectroscopic analyses and
9
X-ray diffraction method.19 The H and 31P NMR data of 6, as
1
well as those of I and II, are given in Table 3. Undoubtedly, the
spectral data closely resemble each other, in spite of the fact that
these compounds incorporate quite different X ligands. These
results therefore suggest that 6 is structurally analogous to
complexes I and II, in which the molybdenum is eight-
coordinate.
10 S. Kuramochi, Y.-h. Kido, S. Shioda, M. Minato, M. Kakeya,
K. Osakada, Bull. Chem. Soc. Jpn. 2010, 83, 165.
11 a) K. E. Oglieve, R. A. Henderson, J. Chem. Soc., Dalton
Trans. 1991, 3295. b) E. B. Lobkovskii, V. D. Makhaev, A. P.
Borisov, K. N. Semenenko, M. Y. Antipin, Y. T. Struchkov,
Koord. Khim. 1985, 11, 983.
At present, the by-product from the reaction between 3 and a
chlorosilane has not been isolated cleanly. Both the 1H and
31P NMR spectra of the reaction solution are significantly more
complicated than those of 6, and the data are insufficient for
unequivocally assigning the structure of the product. More
detailed studies of the structure are in progress and will be
reported in due course.
12 S. Kuramochi, M. Minato, unpublished results.
1
13 The H and 29Si NMR spectra are consistent with those of a
commercial sample (Shin-Etsu Chemical).
14 D. Astruc, Organometallic Chemistry and Catalysis, Springer,
Berlin, 2007, p. 187.
15 D. Astruc, Organometallic Chemistry and Catalysis, Springer,
Berlin, 2007, p. 184.
In summary, the method reported herein represents a novel
approach to Si-Si bond formation. Bulky chlorosilanes such as
Ph2MeSiCl and Ph3SiCl are substantially unreactive with 1.
However, it has been discovered that their reactions with 2 and 3
proceed in a clean manner to afford the coupling products, which
is probably because of the strong trans-influence of the Si ligand
in these complexes.21
16 a) H. Yamashita, T. Hayashi, T.-a. Kobayashi, M. Tanaka, M.
Goto, J. Am. Chem. Soc. 1988, 110, 4417. b) H. Yamashita,
T.-a. Kobayashi, T. Hayashi, M. Tanaka, Chem. Lett. 1990,
1447. c) A. A. Zlota, F. Frolow, D. Milstein, J. Chem. Soc.,
Chem. Commun. 1989, 1826.
17 H.-G. Woo, T. D. Tilley, J. Am. Chem. Soc. 1989, 111, 3757.
18 The reaction was undertaken in the presence of 1 (1.0 mmol)
and PhMe2SiCl (4.0 mmol) in THF (20 mL). After stirring at
room temperature for 24 h under argon, the reaction mixture
was treated with H2O and aqueous Na2CO3, and then extracted
with hexane. The extract was washed with brine and dried over
MgSO4. Evaporation of the solution and crystallization from
hexane yielded the product (75% from 1), which was found
by 1H and 29Si NMR spectroscopy to be the previously
characterized PhMe2SiSiMe2Ph.
19 M. Minato, D.-Y. Zhou, K.-i. Sumiura, Y. Oshima, S. Mine,
T. Ito, M. Kakeya, K. Hoshino, T. Asaeda, T. Nakada, K.
Osakada, Organometallics 2012, 31, 4941.
20 H. K. Sharma, K. H. Pannell, Angew. Chem., Int. Ed. 2009, 48,
7052.
21 The incorporation of a silyl group as part of the ligand
framework could result in a number of unique properties for a
metal, see: a) J. Takaya, N. Iwasawa, J. Am. Chem. Soc. 2008,
130, 15254. b) M. C. MacInnis, R. McDonald, M. J. Ferguson,
S. Tobisch, L. Turculet, J. Am. Chem. Soc. 2011, 133, 13622.
c) Y. Lee, J. C. Peters, J. Am. Chem. Soc. 2011, 133, 4438.
References and Notes
1
2
M. A. Brook, Silicon in Organic, Organometallic, and
Polymer Chemistry, John Wiley & Sons, New York, 2000,
Chap. 11.
For reviews of this topic, see: a) R. West, in Comprehensive
Organometallic Chemistry II, ed. by A. G. Davies, Pergamon,
Oxford, 1995, Vol. 2, p. 77. doi:10.1016/B978-008046519-
7.00015-0. b) H. Yamashita, M. Tanaka, Bull. Chem. Soc. Jpn.
1995, 68, 403. c) R. West, in The Chemistry of Organic
Silicon Compounds, ed. by Z. Rappoport, Y. Apeloig, John
Wiley & Sons, New York, 2001, Vol. 3, Chap. 9. doi:10.1002/
0470857269.ch9. d) G. M. Gray, J. Y. Corey, in Silicon-
Containing Polymers: The Science and Technology of Their
Synthesis and Applications, ed. by R. G. Jones, W. Ando, J.
Chojnowski, Kluwer Academic Publishers, Dordrecht, 2000,
Chap. 14. doi:10.1007/978-94-011-3939-7_14.
3
See for example: a) J. Koe, Polym. Int. 2009, 58, 255. b) A.
Saxena, K. Okoshi, M. Fujiki, M. Naito, G. Guo, T. Hagihara,
Chem. Lett. 2014, 43, 1005–1007 | doi:10.1246/cl.140203
© 2014 The Chemical Society of Japan | 1007