Synthesis and structure of a hydrido(hydrosilylene)ruthenium complex and its reactions with nitriles

Article abstract of DOI:10.1002/anie.200703154

(Chemical Equation Presented) Ring around the Ru=Si: Abstraction of a pyridine ligand with BPh₃ was used to synthesize the silylene complex 1 (see scheme), which reacts with nitriles at room temperature to give silylisocyanide complex 3 by C-C bond activation. The structure of key intermediate 2 with an η²-Si-H agostic interaction is given along with mechanisms for the formation of 2 and 3.

Full text of DOI:10.1002/anie.200703154

Communications
DOI: 10.1002/anie.200703154
Silylene Chemistry
Synthesis and Structure of a Hydrido(hydrosilylene)ruthenium
Complex and Its Reactions with Nitriles**
Mitsuyoshi Ochiai, Hisako Hashimoto,* and Hiromi Tobita*
Transition-metal–silylene complexes have attracted much
attention during the last two decades, as they are considered
key intermediates in various transformation reactions of
organosilicon compounds.[1] Numerous silylene complexes
have been prepared and their reactivities investigated, but
clear-cut reactions of silylene complexes with substrates have
been limited, because two bulky groups on the silylene ligand,
which have been necessary to stabilize the silylene complexes,
only allowed access of very small substrate molecules to the
reactive silylene silicon site. Recently, Tilley et al.[2] and our
group[3] have succeeded in synthesizing the silylene complexes
having a hydrogen atom on the silylene ligand, and demon-
strated that these complexes can react cleanly even with
normal-sized substrate molecules such as ketones, oxiranes,
and nitriles. These results imply that silylene complexes
(CO)(py)Me][4] with H3SiC(SiMe3)3. Abstraction of pyridine
with BPh3 occurred immediately, and complex 1 was obtained
1
in 71% yield as orange crystals. The H NMR spectrum of 1
has a signal for the SiH hydrogen at d = 9.14 ppm, which is in
a typical chemical shift range for the hydrogen atom on an sp2
silicon atom (d = 6.7–12.1 ppm).[2b,c,3a,5] A resonance arising
from the RuH hydrogen appears at d = À11.19 ppm with no
satellite signals, in contrast to that of WH for 2, which has
satellite signals with 2JSiH = 28.6 Hz.[3a] This indicates that,
unlike 2, complex 1 does not have strong interligand
interaction between the silylene and hydrido ligands.[6]
À
having a less sterically hindered and more reactive Si H bond
on the silylene ligand could develop into a new rich field in
the chemistry of silylene complexes. Herein we report the
synthesis of the first neutral hydrido(hydrosilylene)ruthenium
=
complex [Cp*(CO)(H)Ru Si(H){C(SiMe3)3}] (1, Cp* =
C5Me5) utilizing a new synthetic approach, the full structural
The molecular structure of 1 was determined by X-ray
characterization of 1, and C C bond activation of nitriles by
crystallography (Figure 1).[7] The Ru1 Si1 bond (2.220(2) Š)
À
À
this complex through intermediate formation of an agostic
is the shortest of the base-free silylene ruthenium complexes
2
À
complex with h -Si H coordination.
We reported the synthesis of a tungsten hydrido(hydro-
=
silylene) complex [Cp*(CO)2(H)W Si(H){C(SiMe3)3}] (2) b y
a photoreaction.[3a] Thus, a similar method was applied for the
synthesis of the ruthenium analogue 1: a C6D6 solution of
[Cp*Ru(CO)2Me] and H3SiC(SiMe3)3 was irradiated with UV
light at about 58C. However, instead of 1, silyl complex
[Cp*(CO)2Ru{SiH2C(SiMe3)3}] (3) formed in 76% yield as
determined by NMR spectroscopy. Conversion of 3 into 1 by
removal of CO was unsuccessful even on repeated irradi-
ation–degassing cycles.
Synthesis of 1 was finally achieved by the action of BPh3
on [Cp*(CO)(py)Ru{SiH2C(SiMe3)3}] [4, py = pyridine,
Eq. (1)], which was prepared by the reaction of [Cp*Ru-
Figure 1. ORTEP drawing of 1 (thermal ellipsoids set at 50% proba-
bility). Hydrogen atoms attached to carbon atoms omitted for clarity.
(2.2264–2.33 Š).[8] The sum of the bond angles around the
[*] M. Ochiai, Dr. H. Hashimoto, Prof. H. Tobita
Department of Chemistry
silylene silicon atom is 359(2)8, which is consistent with sp2
À
Graduate Schoolof Science, Tohoku University
Aoba-ku, Sendai 980-8578 (Japan)
Fax: (+81)22-795-6543
E-mail: hhashimoto@mail.tains.tohoku.ac.jp
tobita@mail.tains.tohoku.ac.jp
hybridization. The Si1 H1 bond length is 2.41 Š, which is
À
much longer than those for 2 (1.71(6) Š) and a normal Si H
bond (< 2.0 Š).[6] Clearly, there is no bonding interaction
between them, which is consistent with the above-mentioned
NMR spectroscopy result.
[**] This work was supported by the Ministry of Education, Culture,
Sports, Science and Technology Japan (Grants-in-Aid for Scientific
Research Nos. 18350027, 18064003, 16750044, 19550056, and
19029003.
DFT calculations were carried out to clarify the origin of
the difference in bonding between 1 and 2.[9] In a model
=
complex of 2, [Cp(CO)2(H)W Si(H){C(SiH3)3}] (2’, Cp =
À
C5H5), the direction and symmetry of the W H bonding
orbital is suitable for overlapping with the vacant p orbital of
Supporting Information for this article is available on the WWW
under http://www.angewandte.org or from the author.
8192
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8192 –8194

Angewandte
Chemie
the silylene ligand. In contrast, in a model complex of 1,
bond length is 2.5563(7) Š which is slightly longer than a
[11]
=
À
À
[Cp(CO)(H)Ru Si(H){C(SiH3)3}] (1’), the Ru H bonding
orbital is directed away from the silylene p orbital, and cannot
interact efficiently with it. This difference apparently arises
from the different geometry around the metal center: 1’
adopts a three-legged piano-stool geometry, while 2’ has a
four-legged piano-stool geometry. The details of this theoret-
ical study will be reported elsewhere.
normal Ru Si single bond.
Possible mechanisms for the reaction of 1 with nitriles are
depicted in Scheme 2. This reaction is closely related to the
À
C C bond activation of nitriles by rhodium and iron silyl
The reactivity of 1 toward nitriles was examined. We
reported that the reaction of tungsten complex 2 with nitriles
gave stoichiometric hydrosilylation products.[3b] In contrast,
the reaction of ruthenium complex 1 with nitriles proceeded
À
slowly at room temperature with C C bond cleavage of
nitriles to afford silylisocyanide complexes [Cp*(CO)(R)Ru-
{CNSiH2C(SiMe3)3}] (5a: R = Me, 76%, 5b: R = Ph, 73%;
Scheme 1). The 1H NMR spectrum of 5a shows the signals for
Scheme 2. Reaction mechanisms for the reaction of 1 with nitriles.
[Ru]={Cp*Ru(CO)}, R’=C(SiMe3)3. See text for details.
Scheme 1. Reactions of 1 with nitriles.
complexes, giving silylisocyanide complexes, reported
recently by Brookhart et al. and Nakazawa et al.[12] There
are two possible mechanisms for the formation of intermedi-
ate 6 from 1: one possibility proceeds through 1,2-H migra-
tion to give 16-electron silyl complex A, coordination of a
RuMe and SiH at d = 0.49 and 4.71 ppm (1JSiH = 228.0 Hz),
and the 13C NMR spectrum exhibits the signals for CN and
CO at d = 196.8 and 207.0 ppm, respectively. In the IR
spectrum of 5a, the stretching vibrations for CO and CN
are observed at 1923 and 2013 cmÀ1, respectively.
nitrile molecule to ruthenium to give B, silyl ligand migration
2
À
to nitrogen, and h -coordination of an Si H bond. This
Interestingly, at the early stage of this reaction, an
mechanism is analogous to the Brookhart–Nakazawa mech-
=
iminoacyl complex with an agostic interaction at one of the
anism that produces a metal–C N three-membered-ring
À
=
Si H bonds, [Cp*(CO)Ru{C(R) NSiH2C(SiMe3)3}] (6a: R =
Me, 6b: R = Ph), was detected as an intermediate by NMR
iminoacyl complex.[12] In contrast, the other mechanism
starts from the coordination of a nitrile molecule to the
silylene silicon to generate C. Then, [2+2] cycloaddition,
giving four-membered metallacyle D, followed by partial
reductive elimination of the Si H bond forms 6. From 6,
decomplexation of the h -Si H bond and intramolecular C C
bond oxidative addition leads to 5.
spectroscopy (see Scheme 1). Complex 6a shows the signals
2
for the h -coordinated Si H moiety upfield in the 29Si NMR
À
1
À
spectrum (d = À85.2 ppm, 223 K) and also in the H NMR
2
À
À
spectrum (d = À10.53 ppm), and the
large coupling constant 1JSiH
(72.8 Hz) observed in the latter is
To determine which intermediate, B or C, is formed in the
initial step, we monitored the reaction of 1 with pyridine by 1H
and 29Si NMR spectroscopy. The reaction proceeded at
2
À
consistent with h -Si H coordina-
tion (a typical range is 20–
70 Hz).[6,10] The structure of 6a was
confirmed by the X-ray crystal-
structure analysis of the isolated h5-
C5Me4Et derivative, 6c (Experimen-
tal Section, Figure 2).[7] The Si1 atom
is in a tetrahedral environment, not
À508C to afford
a
base-stabilized silylene complex
=
[Cp*(CO)(H)Ru Si(py)(H){C(SiMe3)3}] (7), which has a
29Si NMR signal at d = 98.6 ppm. This chemical shift is
characteristic for base-stabilized silylene complexes. Subse-
quent warming of this mixture to room temperature resulted
in conversion of 7 into silyl complex 4 (Scheme 3). This result
implies that, in the reaction of 1 with nitriles, the initial step is
the formation of base-stabilized silylene complex C. From C,
either [2+2] cycloaddition to form D or conversion into B
could occur to produce 6. However, at present, we cannot
determine conclusively which mechanism is operative.
À
counting the Ru1 Si1 bond, and is
thus sp3-hybridized. The Si1 H2
À
bond length (1.71(3) Š) indicates a
Figure 2. ORTEP drawing
of 6c (thermal ellipsoids
À
significant Si H bond elongation
that suggests substantial Ru (dp) to
set at 50% probability).
Hydrogen atoms
À
Si H (s*) back-donation (the typi-
Importantly, there is another possible reaction from C,
that is, the migration of the hydrido ligand from ruthenium to
the coordinated nitrile carbon to result in hydrosilylation.
2
À
cal h -Si H bond length is in the
attached to carbon
atoms omitted for clarity. range of 1.6–1.9 Š).[10b] The Ru1 Si1
À
Angew. Chem. Int. Ed. 2007, 46, 8192 –8194ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8193

Communications
[1] a) T. D. Tilley in The Chemistry of Organic Silicon Compounds
(Eds.: S. Patai, Z. Rappoport), Wiley, New York, 1989, pp. 1415 –
1477; b) M. S. Eisen in The Chemistry of Organic Silicon
Compounds, Vol. 2 (Eds.: Z. Rappoport, Y. Apeloig), Wiley,
New York, 1998, pp. 2037 – 2128; c) H. Ogino, Chem. Rec. 2002,
2, 291 – 306; d) M. Okazaki, H. Tobita, H. Ogino, Dalton Trans.
2003, 493 – 506.
Scheme 3. Reaction of 1 with pyridine.
[2] a) P. B. Glaser, T. D. Tilley, J. Am. Chem. Soc. 2003, 125, 13640 –
13641; b) B. V. Mork, T. D. Tilley, A. J. Schultz, J. A. Cowan, J.
Am. Chem. Soc. 2004, 126, 10428 – 10440; c) P. G. Hayes, C.
Beddie, M. B. Hall, R. Waterman, T. D. Tilley, J. Am. Chem. Soc.
2006, 128, 428 – 429.
[3] a) T. Watanabe, H. Hashimoto, H. Tobita, Angew. Chem. 2004,
116, 220 – 223; Angew. Chem. Int. Ed. 2004, 43, 218 – 221; b) T.
Watanabe, H. Hashimoto, H. Tobita, J. Am. Chem. Soc. 2006,
128, 2176 – 2177; c) H. Hashimoto, M. Ochiai, H. Tobita, J.
Organomet. Chem. 2007, 692, 36 – 43.
This type of reaction is indeed a dominant process in the
reaction of tungsten complex 2 with nitriles, but surprisingly it
was not observed in the case of ruthenium complex 1. To
clarify the reason for this difference, the mechanistic studies
on the reaction of hydrido(hydrosilylene) complexes with
nitriles and other polar unsaturated organic compounds are in
progress.
[4] M. Iwata, M. Okazaki, H. Tobita, Organometallics 2006, 25,
6115 – 6124.
[5] a) R. S. Simons, J. C. Gallucci, C. A. Tessier, W. J. Youngs, J.
Organomet. Chem. 2002, 654, 224 – 228; b) J. D. Feldman, J. C.
Peters, T. D. Tilley, Organometallics 2002, 21, 4065 – 4075;
c) P. B. Glaser, T. D. Tilley, Organometallics 2004, 23, 5799 –
5812.
Experimental Section
All manipulations were conducted in an atmosphere of dry argon or
nitrogen by employing either standard Schlenk techniques or a
glovebox.
1: Hexane (6 mL) was added to 4 (171 mg, 0.283 mmol) and BPh3
(69 mg, 0.28 mmol) and the mixture was stirred for 4 h at room
temperature. After 4 h, crystals of (py)BPh3 were separated off by
filtration and the filtrate was evaporated under vacuum to give an
orange residue. Pentane (ca. 7 mL) was added to the residue, the
mixture was cooled to À308C, and the precipitate (py)BPh3 was
filtered off by a membrane filter. The filtrate was concentrated under
vacuum, and the residue was recrystallized from pentane (ca. 2 mL) at
À308C to afford 1 as orange crystals in 71% yield (106 mg,
0.202 mmol). 1H NMR (300 MHz, C6D6): d = À11.19 (br s, 1H,
RuH), 0.38 (s, 27H, SiMe3), 1.87 (s, 15H, Cp*), 9.14 ppm (br s,
1JSiH = 136.9 Hz, 1H, SiH). 13C{1H} NMR (75.5 MHz, C6D6): d = 4.2
(SiMe3), 11.1 (C5Me5), 29.2 (C(SiMe3)3), 79.5 (C5Me5), 210.4 ppm
(CO). 29Si{1H} NMR (59.6 MHz, C6D6): d = À4.1 (SiMe3), 337.5 (SiH).
IR (KBr pellet): n˜ = 2021 (m, nSiH), 1973 (vs, nCO), 1923 cmÀ1 (s, nRuH).
Elemental analysis (%) calcd for C21H44OSi4Ru: C 47.95, H 8.43;
found: C 47.82, H 8.60.
[6] J. Y. Corey, J. Braddock-Wilking, Chem. Rev. 1999, 99, 175 – 292.
[7] 1: orthorhombic, P212121; a = 11.3837(3), b = 13.7325(6), c =
17.3339(7) Š, V= 2709.7(2) Š3, Z = 4; C21H44OSi4Ru, T=
150(2) K, 23179 reflections, 6038 independent reflections
(Rint = 0.0894), R1 = 0.0621 (I > 2s(I)), wR2 = 0.1110; m =
0.765 mmÀ1. 6c: triclinic; P1; a = 9.1065(4), b = 11.8594(5), c =
¯
14.6238(8) Š, a = 72.630(3), b = 85.785(3), g = 81.268(2)8, Z = 2;
C24H49NOSi4Ru, T= 150(2) K, 13827 reflections, 6572 independ-
ent reflections (Rint = 0.0302), R1 = 0.0289 (I > 2s(I)), wR2 =
0.0975; m = 0.704 mmÀ1; refinement by full-matrix least-squares
2
À
À
methods on F . The positions of the Si H and the Ru H
hydrogen atoms were located in the difference Fourier electron-
density map and were refined with isotropic thermal parameters.
CCDC-653361 (1) and 653364 (6c) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
6c: As soon as acetonitrile (8.8 mL, 0.17 mmol) was added to a
[8] a) S. K. Grumbine, G. P. Mitchell, D. A. Straus, T. D. Tilley, A. L.
Rheingold, Organometallics 1998, 17, 5607 – 5619; b) T. A.
Schmedake, M. Haaf, B. J. Paradise, A. J. Millevolte, D. R.
Powell, R. West, J. Organomet. Chem. 2001, 636, 17 – 25; c) D.
Amoroso, M. Haaf, G. P. A. Yap, R. West, D. E. Fogg, Organo-
metallics 2002, 21, 534 – 540.
[9] The calculations were performed with Gaussian98 Rev. A.11,
M. J. Frisch et al. (see Supporting Information), Gaussian, Inc.,
Pittsburgh, PA, 1998. Geometry optimizations were performed
at B3LYP level by using 6-31(G)d for H, C, O and Si atoms and
LANL2DZ for Ru or W atoms.
[10] a) U. Schubert, Adv. Organomet. Chem. 1990, 30, 151 – 187; b) Z.
Lin, Chem. Soc. Rev. 2002, 31, 239 – 245; c) S. Lachaize, S. Sabo-
Etienne, Eur. J. Inorg. Chem. 2006, 2115 – 2127.
[11] Based on a survey of the Cambridge Structural Database, CSD
version 5.27 (November 2005).
5
=
hexane solution of [(h -C5Me4Et)(CO)(H)Ru Si(H){C(SiMe3)3}]
(1b; 61 mg, 0.11 mmol), volatile components were removed under
vacuum. Hexane (1 mL) was added to the residue, and toluene was
added dropwise until the precipitate dissolved. The solution was
cooled to À308C for two days to give 6c as yellow crystals in 41%
yield (26 mg, 0.045 mmol). 1H NMR (300 MHz, [D8]toluene): d =
1
À10.60 (d, 2JHH = 9.2 Hz, JSiH = 74.7 Hz, 1H, h2-SiH), 0.43 (s, 27H,
3
SiMe3), 0.83 (t, JHH = 7.7 Hz, 3H, CH2CH3) 1.60 (s, 6H, C5Me4Et),
3
1.65 (s, 6H, C5Me4Et), 2.12 (q, JHH = 7.7 Hz, 2H, CH2CH3), 2.44 (s,
2
1
3H, NCMe), 4.80 ppm (d, JHH = 9.2 Hz, JSiH = 225.0 Hz, 1H, SiH).
29Si{1H} NMR (59.6 MHz, [D8]toluene, 243 K): d = À86.5 (SiH),
À0.1 ppm (SiMe3). MS (EI, 70 eV) 581 (M+, 32), 566 (M+ÀCH3,
100), 522 (M+ÀCOÀH, 5), 534 (M+À3CH3À2H, 48), 73 (SiMe3, 24).
Elemental analysis (%) calcd for C24H49NOSi4Ru: C 49.61, H 8.50, N
2.41; found: C 49.82, H 8.24, N 2.55.
[12] a) F. L. Taw, P. S. White, R. G. Bergman, M. Brookhart, J. Am.
Chem. Soc. 2002, 124, 4192 – 4193; b) F. L. Taw, A. H. Mueller,
R. G. Bergman, M. Brookhart, J. Am. Chem. Soc. 2003, 125,
9808 – 9813; c) H. Nakazawa, T. Kawasaki, K. Miyoshi, C. H.
Suresh, N. Koga, Organometallics 2004, 23, 117 – 126; d) H.
Nakazawa, K. Kamata, M. Itazaki, Chem. Commun. 2005, 4004 –
4006.
Received: July 14, 2007
Published online: September 21, 2007
Keywords: agostic interactions · C–C activation ·
.
rearrangement · ruthenium · silicon
8194
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8192 –8194