Mechanisms of C-Si and C-H Bond Formation on the Reactions of Alkenylruthenium(II) Complexes with Hydrosilanes

Article abstract of DOI:10.1021/om9909035

Reactions of the four alkenylruthenium(II) complexes Ru[C(R¹)=CH(R²)]Cl(CO)(PPh₃)₂ (R¹ = H, R² = Ph (1b); R¹ = H, R² = t-Bu (1c); R¹ = Ph, R² = Ph (1d); R¹ = CH=CH(SiMe₃), R² = SiMe₂Ph (1e)) with HSiMe₂Ph, which constitute the product-forming step of ruthenium-catalyzed hydrosilylation of alkynes, have been examined. Two reaction courses are operative: one provides the C-Si coupling product PhMe₂SiC(R¹)=CH(R₂) and RuHCl(CO)-(PPh3)3 (path A), and the other forms the C-H coupling product HC(R¹)=CH(R²) and Ru(SiMe₂Ph)Cl(CO)(PPh₃)₂ (path B). The ratio of the two courses significantly varies with substituents on the alkenyl ligands, particularly with the α-substituent (R¹). Thus, 1b and 1c, without an α-substituent, react mainly by path A. In contrast, 1d and 1e, bearing an α-substituent, exclusively undergo path B. Kinetic studies using 1b and its para-substituted styryl ligand derivatives have revealed that path A proceeds by direct interaction of the five-coordinated complexes with hydrosilane, without dissociation of the PPh₃ ligand. On the other hand, path B involves dissociation of PPh₃ prior to the reaction of 1d or 1e with hydrosilane. Mechanisms of the C-Si and C-H bond formation are discussed with kinetic data in detail.

Full text of DOI:10.1021/om9909035

1308
Organometallics 2000, 19, 1308-1318
Mech a n ism s of C-Si a n d C-H Bon d F or m a tion on th e
Rea ction s of Alk en ylr u th en iu m (II) Com p lexes w ith
Hyd r osila n es
Yooichiroh Maruyama, Kunihiro Yamamura, Takashi Sagawa,
Hiroyuki Katayama, and Fumiyuki Ozawa*
Department of Applied Chemistry, Faculty of Engineering, Osaka City University,
Sumiyoshi-ku, Osaka 558-8585, J apan
Received November 15, 1999
Reactions of the four alkenylruthenium(II) complexes Ru[C(R1)dCH(R2)]Cl(CO)(PPh3)2 (R1
) H, R2 ) Ph (1b); R1 ) H, R2 ) t-Bu (1c); R1 ) Ph, R2 ) Ph (1d ); R1 ) CHdCH(SiMe3), R2
) SiMe2Ph (1e)) with HSiMe2Ph, which constitute the product-forming step of ruthenium-
catalyzed hydrosilylation of alkynes, have been examined. Two reaction courses are
operative: one provides the C-Si coupling product PhMe2SiC(R1)dCH(R2) and RuHCl(CO)-
(PPh3)3 (path A), and the other forms the C-H coupling product HC(R1)dCH(R2) and
Ru(SiMe2Ph)Cl(CO)(PPh3)2 (path B). The ratio of the two courses significantly varies with
substituents on the alkenyl ligands, particularly with the R-substituent (R1). Thus, 1b and
1c, without an R-substituent, react mainly by path A. In contrast, 1d and 1e, bearing an
R-substituent, exclusively undergo path B. Kinetic studies using 1b and its para-substituted
styryl ligand derivatives have revealed that path A proceeds by direct interaction of the
five-coordinated complexes with hydrosilane, without dissociation of the PPh3 ligand. On
the other hand, path B involves dissociation of PPh3 prior to the reaction of 1d or 1e with
hydrosilane. Mechanisms of the C-Si and C-H bond formation are discussed with kinetic
data in detail.
Sch em e 1
In tr od u ction
Catalytic hydrosilylation of alkenes and alkynes is a
versatile synthetic means of obtaining organosilicon
compounds.1 This catalysis is generally thought to
proceed by either the Chalk-Harrod mechanism2 or the
modified Chalk-Harrod mechanism.3 Although there
are several variations,4,5 the essential features of the
catalytic cycles can be summarized as depicted in
Scheme 1.6 The former mechanism (cycle CH) invokes
(1) (a) Hiyama, T.; Kusumoto, T. In Comprehensive Organic Syn-
thesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford, U.K.,
1991; Vol. 8, p 763. (b) Ojima, I. In The Chemistry of Organic Silicon
Compounds; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, U.K.,
1989; p 1479.
(2) (a) Chalk, A. J .; Harrod, J . F. J . Am. Chem. Soc. 1965, 87, 16.
(b) Harrod, J . F.; Chalk, A. J . J . Am. Chem. Soc. 1965, 87, 1133.
(3) (a) Seitz, F.; Wrighton, M. S. Angew. Chem., Int. Ed. Engl. 1988,
27, 289. (b) Randolph, C. L.; Wrighton, M. S. J . Am. Chem. Soc. 1986,
108, 3366. (c) Ferna´ndez, M. J .; Esteruelas, M. A.; J ime´nez, M. S.; Oro,
L. A. Organometallics 1986, 5, 1519. (d) Ojima, I.; Fuchikami, T.;
Yatabe, M. J . Organomet. Chem. 1984, 260, 335. (e) Onopchenko, A.;
Sabourin, E. T.; Beach, D. L. J . Org. Chem. 1983, 48, 5101. (f)
Schroeder, M. A.; Wrighton, M. S. J . Organomet. Chem. 1977, 128,
345.
(4) (a) Stein, J .; Lewis, L. N.; Gao, Y.; Scott, R. A. J . Am. Chem.
Soc. 1999, 121, 3693. (b) LaPointe, A. M.; Rix, F. C.; Brookhart, M. J .
Am. Chem. Soc. 1997, 119, 906. (c) Esteruelas, M. A.; Olivan, M.; Oro,
L. A. Organometallics 1996, 15, 814. (d) Maddock, S. M.; Rickard, C.
E. F.; Roper, W. R.; Wright, L. J . Organometallics 1996, 15, 1793. (e)
Hofmann, P.; Meier, C.; Hiller, W.; Heckel, M.; Riede, J .; Schmidt, M.
U. J . Organomet. Chem. 1995, 490, 51. (f) Hostetler, M. J .; Butts, M.
D.; Bergman, R. Organometallics 1993, 12, 65. (g) Esteruelas, M. A.;
Herrero, J .; Oro, L. A. Organometallics 1993, 12, 2377. (h) Bergens,
S. H.; Noheda, P.; Whelan, J .; Bosnich, B. J . Am. Chem. Soc. 1992,
114, 2128. (i) Duckett, S. B.; Pertsz, R. N. Organometallics 1992, 11,
90. (j) Tanke, R. S.; Crabtree, R. H. Organometallics 1991, 10, 415. (k)
Lewis, L. N. J . Am. Chem. Soc. 1990, 112, 5998. (l) Hostetler, M. J .;
Bergman, R. J . Am. Chem. Soc. 1990, 112, 8621.
insertion of alkene (or alkyne) into a metal-hydrogen
bond (process i), whereas the latter (cycle mCH) invokes
insertion into a metal-silicon bond (process iii). Hy-
drosilylation products are formed by subsequent C-Si
and C-H bond formation reactions (processes ii and iv,
10.1021/om9909035 CCC: $19.00 © 2000 American Chemical Society
Publication on Web 03/03/2000

Reactions of Ru Alkenyl Complexes with Hydrosilanes
Organometallics, Vol. 19, No. 7, 2000 1309
Sch em e 2
respectively), typically between alkyl (or alkenyl) inter-
mediates (B and D) and hydrosilanes. These mecha-
nisms have accounted for a number of observations
associated with catalytic hydrosilylation. Furthermore,
their experimental supports based on the studies of
presumed intermediates have been presented in recent
years.4-6 However, the reason for alteration of the
catalytic mechanism depending upon catalysts and
substrates has not been well-understood, though such
information is crucial for designing highly selective
catalyses.7
We have recently found that the selection of one of
the two catalytic cycles in ruthenium-catalyzed hydrosi-
lylation of 1-(trimethylsilyl)-1-buten-3-yne takes place
at a product-forming step, which is composed of the
reaction of butadienylruthenium intermediate 1a in
Scheme 1 with hydrosilane.8 Complex 1a corresponds
to intermediate B in cycle CH. When the C-Si bond
formation affording 4a (process ii) proceeds, a ruthe-
nium hydride simultaneously formed in the catalytic
system starts the Chalk-Harrod cycle. On the other
hand, when the C-H bond formation giving 5a takes
place according to process v, a silylruthenium species
generated in the system undergoes insertion of 1-(tri-
methylsilyl)-1-buten-3-yne to start the modified Chalk-
Harrod cycle. Therefore, we have been interested in the
mechanisms of the C-Si and C-H bond formation
reactions.
There are two possible mechanisms for the reaction
of organotransition-metal complexes with hydrosilanes.
One is a sequence of oxidative addition and reductive
elimination, and the other is σ-bond metathesis involv-
ing a four-center transition state. The former process
is commonly assumed for late-transition-metal-cata-
lyzed reactions.2-4,9 On the other hand, the latter has
been documented for early-transition-metal complexes
with a d0 metal center.5a,b,10 Recently, the possibility of
σ-bond metathesis has been suggested also for late-
transition-metal systems mainly by theoretical stud-
ies,11 while its experimental evidence is still limited.12,13
Therefore, we employed in this study four alkenylru-
thenium complexes which can be isolated (1b-e in
Scheme 2). Kinetic studies have revealed that clearly
different mechanisms are operative for the C-Si and
C-H bond formation, respectively. Thus, the C-Si bond
formation proceeds by direct interaction of the five-
coordinated complex with hydrosilane without dissocia-
tion of PPh3 ligand, while the C-H bond formation
involves prior dissociation of PPh3.
Resu lts a n d Discu ssion
Rea ction s of Alk en ylr u th en iu m Com p lexes w ith
HSiMe2P h . The reactions of 1b-e with HSiMe2Ph in
CDCl3 (Scheme 2) were examined by NMR spectroscopy
and GC-mass spectrometry. Table 1 summarizes the
results. All of the reactions proceeded at 30 °C by two
reaction courses (path A and path B), giving the C-Si
coupling product 4 and the C-H coupling product 5,
respectively. The selectivities listed in the table are
based on relative ratios of 4 to 5 determined by 1H NMR
spectroscopy. The formation of 4 by path A was ac-
companied by generation of a hydridoruthenium com-
plex, which was identified as RuHCl(CO)(PPh3)3 (2)
when the reaction was carried out in the presence of
added PPh3. However, since the hydride complex is in
need of three PPh3 ligands to be stabilized while the
starting 1 possesses only two PPh3 ligands, the hydride
complex could not be detected in the absence of added
In this paper we wish to report our mechanistic
studies on the reactions of alkenylruthenium(II) com-
plexes with hydrosilanes.14 Since 1a in Scheme 1 is too
unstable to be isolated, further study on the mecha-
nisms of C-Si and C-H bond formation was infeasible.
(5) (a) Fu, P. E.; Brard, L.; Li, Y.; Marks, T. J . J . Am. Chem. Soc.
1995, 117, 7157. (b) Kesti, M. R.; Waymouth, R. M. Organometallics
1992, 11, 1095. (c) Takahashi, T.; Hasegawa, M.; Suzuki, N.; Saburi,
M.; Rousset, C. J .; Fanwick, P. E.; Negishi, E. J . Am. Chem. Soc. 1991,
113, 8564.
(6) (a) Brunstein, P.; Knorr, M. J . Organomet. Chem. 1995, 500, 21.
(b) Tilley, T. D. In The Chemistry of Organic Silicon Compounds; Patai,
S., Rappoport, Z., Eds.; Wiley: Chichester, U.K., 1989; p 1415. (c)
Recatto, C. A. Aldrichim. Acta 1995, 28, 85.
(7) The relative ease of the Chalk-Harrod and modified Chalk-
Harrod mechanisms has been theoretically examined: (a) Sakaki, S.;
Mizoe, N.; Sugimoto, M. Organometallics 1998, 17, 2510. (b) Sakaki,
S.; Ogawa, M.; Musashi, Y.; Arai, T. J . Am. Chem. Soc. 1994, 116,
7258. (c) Sugimoto, M.; Yamasaki, I.; Mizoe, N.; Anzai, M.; Sakaki, S.
Theor. Chem. Acc. 1999, 102, 377-384.
(8) Maruyama, Y.; Yamamura, K.; Nakayama, I. Yoshiuchi, K.
Ozawa, F. J . Am. Chem. Soc. 1998, 120, 1421.
(9) (a) Aizenberg, M.; Milstein, D. J . Am. Chem. Soc. 1995, 117,
6456. (b) Van der Boom, M. E.; Ott, J .; Milstein, D. Organometallics
1998, 17, 4263.
(11) (a) Maseras, F.; Duran, M.; Lledo´s, A.; Bertra´n, J . J . Am. Chem.
Soc. 1992, 114, 2922. (b) Musaev, D. G.; Mebel, A. M.; Morokuma, K.
J . Am. Chem. Soc. 1994, 116, 10693. (c) Siegbahn, P. E. M.; Crabtree,
R. H. J . Am. Chem. Soc. 1996, 118, 4442. (d) Su, M.-D.; Chu, S.-Y. J .
Am. Chem. Soc. 1997, 119, 5373. (e) Hutschka, F.; Dedieu, A.;
Eichberger, M.; Fornika, R.; Leitner, W. J . Am. Chem. Soc. 1997, 119,
4432. (f) Milet, A.; Dedieu, A.; Kapteijn, G.; van Koten, G. Inorg. Chem.
1997, 36, 3223.
(12) (a) Luo, X.-L.; Crabtree, R. H. J . Am. Chem. Soc. 1989, 111,
2527. (b) Hartwig, J . F.; Bhandari, S.; Rablen, P. R. J . Am. Chem. Soc.
1994, 116, 1839. (c) Iverson, C. N.; Smith, M. R., III. J . Am. Chem.
Soc. 1995, 117, 4403. (d) Lee, J . C., J r.; Peris, E.; Rheingold, A. L.;
Crabtree, R. H. J . Am. Chem. Soc. 1994, 116, 11014. (e) Marder, T.
B.; Norman, N. C.; Rice, C. R.; Robins, E. G. Chem. Commun. 1997,
53.
(10) (a) Voskoboynikov, A. Z.; Parshina, I. N.; Shestakova, A. K.;
Butin, K. P.; Beletskaya, I. P.; Kuz’mina, L. G.; Howard, J . A. K.
Organometallics 1997, 16, 4041. (b) Radu, N. S.; Tilley, T. D. J . Am.
Chem. Soc. 1995, 117, 5863. (c) Woo, H.-G.; Walzer, J . F.; Tilley, T. D.
J . Am. Chem. Soc. 1992, 114, 7047.
(13) For reviews on σ-bond metathesis, see: (a) Arndtsen, B. A.;
Bergman, R. G.; Mobley, T. A.; Peterson, T. H. Acc. Chem. Res. 1995,
28, 154. (b) Crabtree, R. H. Chem. Rev. 1995, 95, 987.
(14) Portions of this work have been communicated: Maruyama,
Y.; Yamamura, K.; Ozawa, F. Chem. Lett. 1998, 905.

1310 Organometallics, Vol. 19, No. 7, 2000
Maruyama et al.
Ta ble 1. Rea ction s of 1b-e w ith HSiMe2P h (20
Ta ble 2. P seu d o-F ir st-Or d er Ra te Con sta n ts for
th e Rea ction of 1b w ith HSiMe2P h u n d er Va r iou s
Con d ition sa
equ iv)a
selectivity (%)c
[HSiMe2Ph] [PPh3]
run
complex
PPh3b (equiv/1)
path A
path B
run
(mM)
(mM)
reacn temp (°C) 103kobsd (s-1
)
1
2
3
4
5
6
7
8
1b
1b
1c
1c
1c
1c
1d
1d
1e
1e
0
1
0
0.5
1
2
0
1
0
100
99
67
78
91
93
0
0
0
0
0
1
33
22
9
1
2
3
4
5
6
7
8
82
99
10
10
10
10
10
0
20
30
10
10
10
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
0.0
0.550(5)
0.66(2)
1.06(2)
1.34(1)
2.02(6)
1.5(1)
1.26(3)
1.18(1)
0.738(9)
2.32(4)
3.86(5)
151
200
285
200
200
200
200
200
200
7
100
100
100
100
9
9
10
11
10d
1
20.0
30.0
a
All reactions were run in CDCl3 at 30 °C. b The amount of PPh3
initially added to the system. c The selectivity was based on the
a
All reactions were run in CDCl3. Initial concentration: [1b]0
ratio of 4 to 5, determined by 1H NMR spectroscopy. The reaction
d
) 10.0 mM.
also gave a small amount (ca. 4%) of PhMe2SiCtCCHdCHSiMe3,
which is not the reaction product with HSiMe2Ph but a â-hydrogen
elimination product from 1e.8
Sch em e 3
PPh3 due to decomposition. On the other hand, when
path B was operative, corresponding amounts of silyl
complex 3 were formed in the reaction systems irrespec-
tive of the presence or absence of added PPh3. The
C-H coupling product formed from 1e (runs 9 and 10)
was a 9:1 mixture of the two structural isomers PhMe2-
SiCHdCdCHCH2SiMe3 (5e′) and PhMe2SiCHdCHCHd
CHSiMe3 (5e′′). As already mentioned in the previous
paper,8 the minor product 5e′′ is formed directly from
1e, while the major product 5e′ is produced via a 1,3-
shift of the RuCl(CO)(PPh3)2 moiety along the 2-buta-
dienyl ligand in 1e.
As seen from Table 1, the ratio of path A to path B
significantly varied with substituent(s) on the alkenyl
ligands, particularly with R-substituent (R1). Thus, 1b
and 1c, without the R-substituent, reacted mainly by
path A (runs 1-6). On the other hand 1d and 1e,
bearing an R-substituent, reacted selectively by path B
(runs 7-10). The selectivities for 1b and 1d were little
affected by addition of PPh3 to the systems. In contrast,
in the reaction of 1c, the ratio of path A to path B
increased as the amount of added PPh3 increased (runs
3-6). These results indicate a steric control of the
reaction course. In the following sections, we first
examine the mechanisms of path A and path B by
kinetic experiments and then discuss the reason for the
change of reaction course depending on the alkenyl
ligands.
in eq 1. On the other hand, the reaction rate was little
dependent upon the concentration of free PPh3 (runs 4
and 6-8).
-d[1b]/dt ) k2nd[HSiMe2Ph]1.04(2)[1b]
k2nd ) [7.0(1)] × 10-3 s-1 M-1 at 10 °C
(1)
These kinetic data indicate the reaction process
depicted in Scheme 3. The first step is the coordination
of hydrosilane to styryl complex 1b.15 Considering a
small influence of added PPh3, this step is assumed to
proceed mainly without dissociation of the PPh3 ligand.
The resulting 6b affords the C-Si coupling product 4b
and RuHCl(CO)(PPh3)2. In the presence of added PPh3,
the latter rapidly combines with PPh3 to give 2.
Kin etic Stu d ies on th e Mech a n ism of P a th A.
Styryl complex 1b was treated with an excess amount
of HSiMe2Ph (>8.2 equiv) in CDCl3 in the presence or
absence of added PPh3, and the time course was followed
by 1H NMR spectroscopy. In the presence of added PPh3,
the reaction was first-order in the concentration of 1b
over 90% conversion. On the other hand, since the
reaction in the absence of added PPh3 was accompanied
by decomposition of a hydridoruthenium species (vide
supra), the reaction rate was estimated at low conver-
sion of 1b (up to 44%), where the first-order plot
exhibited a good linear correlation, including the origin.
There are two possible mechanisms for the formation
of 4b and RuHCl(CO)(PPh3)2 from 6b (Scheme 4). One
is the σ-bond metathesis involving the four-center
transient species 7b (mechanism I), while the other
proceeds by oxidative addition of coordinated hydrosi-
lane, followed by C-Si reductive elimination from the
resulting hydrido-silyl-styryl species 8b (mechanism
II). Mechanism I can take place directly from 6b in least
motion. On the other hand, since 6b is coordinatively
saturated, the oxidative addition in mechanism II must
involve prior dissociation of one of the PPh3 ligands.
Although this situation does not simply agree with the
kinetic data, showing little retardation effect of added
Table 2 lists the rate constants measured under
various conditions. The reaction rate increased linearly
as the concentration of HSiMe2Ph increased (runs 1-5,
r2 ) 0.998). The relation observed was in good agree-
ment with the ideal second-order rate expression given
(15) While the coordination mode of hydrosilane is not defined in
this paper, the η2 coordination of hydrosilane seems the most probable
one. Such structures have been established for many transition-metal
complexes.16

Reactions of Ru Alkenyl Complexes with Hydrosilanes
Organometallics, Vol. 19, No. 7, 2000 1311
Sch em e 4
F igu r e 1. Hammett plot for the data in Table 3.
Ta ble 4. P seu d o-F ir st-Or d er a n d Secon d -Or d er
Ra te Con sta n ts for th e Rea ction s of 1b w ith
Va r iou s Hyd r osila n esa
Ta ble 3. P seu d o-F ir st-Or d er a n d Secon d -Or d er
Ra te Con sta n ts for th e Rea ction s of
Ru {CHdCH(C6H4Y-p)}Cl(CO)(P P h 3)2 (1f-k ) w ith
HSiMe2P h a
103k2nd
run complex (Y)
σp
σp
103kobsd (s-1
)
(s-1 M-1
)
+
1
2
3
4
5
6
7
1f (OMe)
1g (Me)
1h (F)
1b (H)
1i (Cl)
-0.778 -0.268
-0.311 -0.17
5.1(2)
25(1)
1.38(3)
1.43(3)
0.738(9)
0.470(5)
0.44(1)
0.170(5)
6.89(2)
7.17(2)
3.69(4)
2.35(2)
2.18(4)
-0.073
0.062
0
0
0.114
0.15
0.612
0.227
0.232
0.54
1j (Br)
1k (CF3)
0.850(3)
a
All reactions were run in CDCl3 at 0.0 °C in the presence of
added PPh3. Initial concentration: [1]0 ) 10.0 mM, [HSiMe2Ph]0
) 200 mM, [PPh3]0 ) 10.0 mM. The second-order rate constants
(k2nd) were obtained by dividing the pseudo-first-order rate
constants (kobsd) by the initial concentration of HSiMe2Ph.
a
All reactions were run in CDCl3 in the presence of added PPh3.
PPh3, mechanism II may still be possible when the
dissociation of PPh3 from 6b takes place simultaneously
with the oxidative addition and the successive C-Si
reductive elimination proceeds rapidly.17 To distinguish
mechanisms I and II, a variety of substituents were
introduced into 1b and HSiMe2Ph, and their effect on
the reaction rate was investigated.
Initial concentration: [1b]0 ) 10.0 mM, [hydrosilane]0 ) 200 mM,
[PPh3]0 ) 10.0 mM. The second-order rate constants (k2nd) were
obtained by dividing the pseudo-first-order rate constants (kobsd
)
b
by the initial concentration of HSiMe2Ph. The reaction was too
slow to perform a kinetic run, which is complete after ca. 55 h at
30 °C.
the styryl ligand (see Scheme 2 for numbering), which
were measured under the same reaction conditions as
run 9 in Table 2. The more electron-releasing substitu-
ent tended to provide the higher reaction rate. Among
Table 3 summarizes the rate constants for 1f-k
bearing a variety of substituents at the para position of
(16) (a) Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789.
(b) Schubert, U. Adv. Organomet. Chem. 1990, 30, 151. (c) Spaltenstein,
E.; Palma, P.; Kreutzer, K. A.; Willoughby, C. A.; Davis, W. M.;
Buchwald, S. L. J . Am. Chem. Soc. 1994, 116, 10308. (d) Luo, X.-L.;
Kubas, G. J .; Bryan, J . C.; Burns, C. J .; Unkefer, C. J . J . Am. Chem.
Soc. 1994, 116, 10312. (e) Luo, X.-L.; Kubas, G. J .; Burns, C. J .; Bryan,
J . C.; Unkefer, C. J . J . Am. Chem. Soc. 1995, 117, 1159. (f) Corey, J .
Y.; Broddock-Wilking, J . Chem. Rev. 1999, 99, 175.
(17) The kinetic expression given in eq 1 is also consistent with the
mechanism involving the rate-determining formation of 6b. However,
this mechanism is incompatible with the kinetic data for a series of
para-substituted HSiMe2Ph and styryl complexes. Thus, in this
situation, the more electron-rich hydrosilane and the more electron-
deficient ruthenium center should provide the higher reaction rate if
steric conditions are similar. However, the reactivity order observed
for a series of para-substituted dimethylphenylsilane derivatives was
apparently opposite to this prospect (Figure 2). In addition, although
the more electron-withdrawing substituent Y at the para position of
the styryl ligand possibly causes the ruthenium center to be more
electron deficient and thereby a higher reaction rate, the reactivity
order actually observed was just the reverse (Figure 1).
+
various Hammett parameters tested, σp values of the
substituents exhibited the most satisfactory Hammett
correlation with the rate constants (Figure 1): F )
-1.07(4), r2 ) 0.993 except for 1h (Y ) F).18 On the other
hand, Hammett correlation between the kobsd and σp
values was low (r2 ) 0.878).19
Reactivities of a series of para-substituted dimeth-
ylphenylsilanes were compared for the reactions with
1b. The results are listed in runs 1-7 of Table 4. The
more electron-withdrawing substituent tended to pro-
(18) The data including 1h : F ) -1.1(1); r2 ) 0.956.
0
(19) The other Hammett parameters tested were σp and σi. The
correlation coefficients (r2) for least-squares calculations are 0.848 and
0.215, respectively.

1312 Organometallics, Vol. 19, No. 7, 2000
Maruyama et al.
Ta ble 5. Activa tion P a r a m eter s a n d Deu ter iu m
Kin etic Isotop e Effect for th e Rea ction s of Styr yl
Com p lexes 1b,f,k w ith HSiMe2P h a ,b
1f (Y ) OMe)
∆Hq ) 12.0(2) kcal mol-1
kH/kD ) 0.93(7)
∆Sq ) -21.9(8) eu
∆Sq ) -38.3(1) eu
∆Sq ) -48.0(2) eu
1b (Y ) H)
∆Hq ) 8.51(2) kcal mol-1
kH/kD ) 1.00(3)
1k (Y ) CF3)
∆Hq ) 6.67(5) kcal mol-1
kH/kD ) 1.3(1)
a
The activation parameters were estimated from Eyring plots
of the second-order rate constants measured in CDCl3 in the
presence of added PPh3. Initial concentration: [1]0 ) 10.0 mM,
[HSiMe2Ph]0 ) 200 mM, [PPh3]0 ) 10.0 mM. Temperature range:
-35.0-0.0 °C for 1f (r2 ) 0.999 for 5 data points), 0.0-30.0 °C for
1b (r2 ) 1.00 for 4 data points), 0.0-35.0 °C for 1k (r2 ) 1.00 for
b
4 data points). The kD values were measured using DSiMe2Ph
(200 mM) in place of HSiMe2Ph at 0.0 °C in CDCl3 in the presence
of added PPh3 (10.0 mM).
F igu r e 2. Hammett plot for the data in runs 1-7 in Table
On the other hand, mechanism II in Scheme 4 was
found to be less compatible with the experimental
results. Since the oxidative addition of hydrosilane
proceeds via a back-donating interaction between a
filled d orbital on the ruthenium and the H-Si σ*
orbital,16a the more electron-rich ruthenium center and
the more electron-deficient hydrosilane will provide the
higher reaction rate if steric conditions are similar. This
prospect is consistent with the Hammett correlation
observed for a series of para-substituted HSiMe2Ph
(Figure 2). However, the electron density of the ruthe-
nium center was found to be insensitive to the para
substituents Y on the styryl ligands. Thus, the styryl
complex derivatives listed in Table 3 exhibited the same
31P{1H} NMR chemical shifts (δ 31.4(2)) and almost the
same ν(CO) frequencies (1924(2) cm-1, except for 1h
(1912 cm-1)). This fact clearly indicates that the Y
groups are situated too far from the ruthenium center
to affect the electron density. Consequently, it is hard
to rationalize the distinct Hammett correlation observed
from Figure 1 by mechanism II.
4.
Sch em e 5
Table 5 lists the kinetic parameters for 1b, 1f, and
1k . All of the reactions exhibited a large negative
entropy, in agreement with the associative mechanism
in Schemes 3 and 5. The magnitude increased signifi-
cantly as the reactivity of the complex decreased (1f >
1b > 1k ). Deuterium kinetic isotope effects observed
with DSiMe2Ph were negligible for 1b and 1f, while a
small isotope effect beyond the experimental error was
noted for 1k with the least reactivity.
These kinetic parameters can be rationalized by
considering variations in the transition-state structures
between 6 and 7 in Scheme 5. As judged from the kH/kD
values, the exact transition state seems very close to 6
for all complexes. However, the clear Hammett correla-
tion in Figure 1 and a distinct tendency for the kinetic
parameters in Table 5 strongly suggest the occurrence
of such variation dependent upon the para substituents
Y. Thus, when the styryl ligand is highly nucleophilic
due to the presence of an electron-releasing substituent,
the nucleophilic attack on the coordinated hydrosilane
may take place with the least activation barrier after
the formation of 6. The relatively small activation
entropy observed for 1f (Y ) OMe) is consistent with
this situation. In this case, elongation of the H-Si bond
in the transition state should be negligible, and there-
fore no kinetic isotope effect can be observed. In
contrast, when an electron-withdrawing substituent
vide the higher reactivity. Except for the data of
dimethyl(p-methoxyphenyl)silane in run 1, the rate
constants exhibited a fair Hammett correlation with the
σp values of substituents (F ) 0.76(11); r2 ) 0.917)
(Figure 2). On the other hand, reactivities of three
silanes in runs 8-10 decreased in the order HSiMe2Ph
> HSiMePh2 . HSiPh3, probably reflecting the increas-
ing bulkiness.
These kinetic observations can be reasonably inter-
preted by mechanism I in Scheme 4. Thus, the Hammett
correlations given in Figures 1 and 2 indicate nucleo-
philic attack of the styryl ligand on the coordinated
hydrosilane to take place. As depicted by the limiting
structure 7(A) in Scheme 5, this process will produce a
carbocation at the para position of the substituent,
which is stabilized by the resonance structure 7(B). The
good Hammett correlation for σp+ values in Figure 1 is
fully consistent with this mechanism; the σp+ values are
known to reflect stabilization of an R-cationic center
generated at the para position of the substituent in the
transition state.20
(20) (a) Inamoto, N. Hammett Soku (The Hammett Rule); Maruzen:
Tokyo, 1983. (b) Brown, H. C.; Okamoto, Y. J . Am. Chem. Soc. 1958,
80, 4979.

Reactions of Ru Alkenyl Complexes with Hydrosilanes
Organometallics, Vol. 19, No. 7, 2000 1313
F igu r e 3. Effect of added PPh3 on the reaction rate of 1d
with HSiMe2Ph in CDCl3 at 45 °C. Initial concentration:
[1d ]0 ) 10.0 mM, [HSiMe2Ph]0 ) 200 mM.
F igu r e 4. Effect of HSiMe2Ph concentration on the
reaction rate of 1d with HSiMe2Ph in CDCl3 at 45 °C in
the presence of added PPh3. Initial concentration: [1d ]0 )
10.0 mM, [PPh3]0 ) 10.0 mM.
such as CF3 in 1k reduces the nucleophilicity of the
styryl ligand, the transition state may somewhat shift
its position from 6 toward 7. In this situation, the more
restricted structure with an elongated H-Si bond leads
to the larger negative entropy and the observable
isotope effect.
solvents (CDCl3, C6D6, THF-d8).22b The resulting 8d
reacts with HSiMe2Ph to give stilbene 5d and a coor-
dinatively unsaturated silyl complex; the latter subse-
quently combines with free PPh3 in the system to afford
silyl complex 3.
Steady-state approximation for the concentration of
8d leads to the equation
Kin etic Stu d ies on th e Mech a n ism of P a th B. We
next carried out kinetic experiments on path B using
1d and 1e.21 Distinct from 1b, both complexes under-
went dissociation of PPh3 prior to the reaction with
HSiMe2Ph.
d[8d ]/dt ) k1[1d ] - k-1[PPh3][8d ] -
k2[HSiMe2Ph][8d ] ) 0 (2)
(a ) With 1d . The reaction of 1d with HSiMe2Ph in
CDCl3 obeyed good pseudo-first-order kinetics with
respect to the concentration of 1d over 90% conversion
when an excess amount of the hydrosilane (>9.3 equiv)
was present. The reaction progress was effectively
suppressed by addition of free PPh3 to the system; a
good linear correlation between 1/kobsd values and
the concentration of added PPh3 (r2 ) 1.000) was
observed (Figure 3). Furthermore, 1/kobsd values varied
in proportion to 1/[HSiMe2Ph] values (r2 ) 0.999)
(Figure 4).22a
Thus
k1[1d ]
[8d ] )
(3)
k-1[PPh3] + k2[HSiMe2Ph]
Consequently, the rate of the consumption of 1d and
the kobsd value are expressed by eqs 4 and 5, respec-
tively, which are fully consistent with the relations
observed from Figures 3 and 4.
These kinetic observations indicate the mechanism
given in Scheme 6. The first step is dissociation of one
of the PPh3 ligands from 1d . This assumption is
supported by the strong retardation effect of added
PPh3. It is also assumed that the dissociation of PPh3
is accompanied by coordination of the R-phenyl group
of alkenyl ligand to the ruthenium center, because
1d is a 16e complex and dissociation of PPh3 without
an incoming ligand leads to a high-energy process
involving a 14e species. Although the solvent-induced
dissociation is possible, no significant variation in
the reaction rates was observed with three kinds of
d[1d ]
dt
-
) k2[HSiMe2Ph][8d ] )
k1k2[HSiMe2Ph]
k-1[PPh3] + k2[HSiMe2Ph]
[1d ] (4)
k-1[PPh3]
1
1
)
+
(5)
kobsd
k1
k1k2[HSiMe2Ph]
Thus, the k-1/k1k2 values, which are estimated from the
slopes of the plots in Figures 3 and 4, are in good
agreement with each other: [3.80(4)] × 104 and [3.61-
(9)] × 104 s. In addition, reciprocals of the intercepts in
these figures, which correspond to the rate constant k1
for the dissociation of PPh3 ligand from 1d , are in good
accord with each other: [3.0(2)] × 10-3 and [2.8(5)] ×
(21) A listing of the rate constants for 1d and 1e is available in the
Supporting Information.
(22) (a) Kinetic parameters for 1d estimated from pseudo-first-order
rate constants in CDCl3 in the presence of 200 mM of HSiMe2Ph are
as follows: ∆Hq ) 14.7(6) kcal mol-1, ∆Sq ) -25(2) eu, kH/kD ) 1.07-
(3) at 30 °C. (b) Pseudo-first-order rate constants measured in three
kinds of solvents in the presence of 200 mM of HSiMe2Ph and 10.0
10-3 s-1
.
(b) With 1e. It was found that the dissociation of
PPh3 from 1d is a slow process.22c Consequently, further
information on the mechanism of path B, particularly
on the interaction between the alkenylruthenium center
and hydrosilane, could not be gained with 1d . In
mM of added PPh3 at 45 °C are as follows: [CDCl3], [4.51(8)] × 10-4
;
[C6D6], [4.16(4)] × 10-4; [THF-d8], [5.48(7)] × 10-4 s-1
. (c) The
dissociation of PPh3 should be the slowest process in Scheme 6, because
the steady-state approximation for 8d requires the relation k-1, k2
.
k1. However, since k-1 > k2 (k-1/k2 ) [1.0(2)] × 102), the rate-
determining step can be assigned to the process for k2.28g

1314 Organometallics, Vol. 19, No. 7, 2000
Maruyama et al.
Sch em e 6
F igu r e 5. Effect of added PPh3 on the reaction rate of 1e
with HSiMe2Ph in CDCl3 at 20 °C. Initial concentration:
[1e]0 ) 10 mM, [HSiMe2Ph]0 ) 195 mM.
Sch em e 7
F igu r e 6. Effect of HSiMe2Ph concentration on the
reaction rate of 1e with HSiMe2Ph in CDCl3 at 20 °C in
the presence of added PPh3. Initial concentration: [1e]0 )
10 mM, [PPh3]0 ) 0.968 mM.
1
contrast, 31P and H NMR signals of the 2-butadienyl
complex 1e in CDCl3 were significantly broadened even
at -20 °C, indicating the occurrence of fluxional behav-
ior involving a rapid dissociation of PPh3. The high
reactivity is attributable to an apparently six-coordinate
structure of 1e, in which the C3dC4 bond of the
2-butadienyl ligand is weakly bonded to ruthenium by
π-coordination23 and facilitates the phosphine dissocia-
tion (see Scheme 7).
The time course of the reaction of 1e with HSiMe2Ph
in CDCl3 was followed by 1H NMR spectroscopy. A good
first-order dependence on the concentration of 1e
was observed in the presence of an excess amount of
HSiMe2Ph (>9.5 equiv/1e). The reaction was much
faster than that of 1d ,24 probably reflecting much
easier dissociation of PPh3 from 1e than 1d . The
participation of phosphine dissociation in the reac-
tion process was further suggested by the following
kinetic data. Thus, the reaction progress was strongly
retarded by addition of free PPh3, and reciprocals of
the rate constants (1/kobsd) were linearly dependent
upon the concentration of added PPh3 (r2 ) 0.995)
(Figure 5). A good linear correlation was also observed
between 1/kobsd and 1/[HSiMe2Ph] values (r2 ) 1.000)
(Figure 6).
Scheme 7 represents our proposed mechanism. As
already pointed out, 1e is in rapid equilibrium with the
coordinatively unsaturated species 8e, which reacts with
HSiMe2Ph to give the C-H coupling products 5e′ and
5e′′ and silyl complex 3. In this scheme, the presence
of intermediate 9e is postulated. This assumption is
essential to rationalize the typical saturation kinetic
behavior observed for the concentration of HSiMe2Ph
(Figure 6).
When the interconversion between 1e and 9e via 8e
is a rapid process, the concentration of 9e at time t is
given by eq 6, where K1 ) k1/k-1 ) [8e][PPh3]/[1e], K2
) k2/k-2 ) [9e]/[8e][HSiMe2Ph], and [Ru(butadienyl)]total
is the sum of the concentrations of 2-butadienyl ruthe-
nium species at time t (i.e. [1e] + [8e] + [9e]).
K1K2[HSiMe2Ph]
[9e] )
[PPh3] + K1 + K1K2[HSiMe2Ph]
(23) Wakatsuki, Y.; Yamazaki, H.; Maruyama, Y.; Shimizu, I. J .
Organomet. Chem. 1992, 430, C60.
[Ru(butadienyl)]total (6)
(24) Rate constants observed in CDCl3 at 5.0 °C in the presence of
200 mM of HSiMe2Ph are as follows: [1e], 1.92 × 10-3 s-1; [1d ], 4.64
× 10-5 s-1
.
Accordingly, the reaction rate and the kobsd value are

Reactions of Ru Alkenyl Complexes with Hydrosilanes
Organometallics, Vol. 19, No. 7, 2000 1315
Sch em e 8
expressed by eqs 7 and 8, respectively.
d
state approximation for 9e is clearly inconsistent with
the kinetic data.25
The next subject is the identification of the structure
of 9e. Similarly to the reaction of 1b, two types of
mechanisms, the σ-bond metathesis and the sequence
of oxidative addition and reductive elimination, are
possible for the reaction of 8e with HSiMe2Ph. Scheme
8 illustrates those producing the major product 5e′. The
first step, which is common to both of the mechanisms,
is the association of HSiMe2Ph with 8e to give 9e(A).15
When σ-bond metathesis in mechanism III is operative,
9e(A) may be assigned to the intermediate 9e because
the subsequent processes must proceed without any
stable or metastable intermediates. On the other hand,
when mechanism IV takes place, 9e(A) should be a
transient species, while the resultant 9e(B) is consistent
with the intermediate 9e, because oxidative addition of
hydrosilane with late-transition-metal complexes is
commonly a low-energy process.26
These possibilities could be discriminated from each
other on the basis of kinetic parameters measured at a
high concentration of HSiMe2Ph (200 mM) in the
absence of added PPh3. As judging from eq 8, the rate
constants observed under such conditions can be ap-
proximated to k3 in Scheme 7. Indeed, kobsd values were
nearly constant ([1.05(6)] × 10-3 s-1 at 0 °C) indepen-
dent of the concentration of hydrosilane (182-294 mM).
The activation parameters thus obtained are as fol-
lows: ∆Hq ) 20.7(7) kcal mol-1, ∆Sq ) +4(3) eu. Since
the σ-bond metathesis in mechanism III is expected to
involve a significantly large negative entropy arising
from a restricted transition state,13 the observed ∆Sq
value is apparently inconsistent with this mechanism.
On the other hand, the activation parameters fall within
the range previously reported for the C-H reductive
elimination from late-transition-metal complexes.28 Fur-
thermore, the deuterium isotope effect observed with
DSiMe2Ph (kH/kD ) 2.99(9)) was similar to the values
-
[Ru(butadienyl)]total ) k3[9e] )
k3K1K2[HSiMe2Ph]
dt
[Ru(butadienyl)]total
[PPh3] + K1 + K1K2[HSiMe2Ph]
(7)
[PPh3]
k3K1K2[HSiMe2Ph] k3K2[HSiMe2Ph]
1
kobsd
1
1
k3
)
+
+
(8)
On the basis of the slopes and intercepts of Figures 5
and 6, the following equations are derived from eq 8.
From Figure 5 ([HSiMe2Ph]const ) 195 mM)
1
) [7.4(3)] × 105 s M-1
(9)
k3K1K2 × 0.195 M
1
1
+
) [2.7(9)] × 102 s
(10)
k3K2 × 0.195 M k3
From Figure 6 ([PPh3]const ) 0.968 mM)
0.968 × 10-3
M
1
+
) 140(1) s M
(11)
(12)
k3K1K2
k3K2
1
k3
) [1.91(8)] × 102 s
Substitution of the 1/k3 value of eq 12 for eq 10 leads to
1/k3K2 ) 15(19) s M. On the basis of this value, 1/k3K1K2
in eq 11 is estimated as [1.3(2)] × 105 s, which is in good
agreement with the value calculated from eq 9 ([1.4(1)]
× 105 s). The rate and equilibrium constants most
consistent with the data in Figures 5 and 6 are as
follows (20 °C): k3 ) [5.2(2)] × 10-3 s-1, K1K2 ) [1.4(4)]
(25) The steady-state approximation for 9e leads to the following
equation, which does not coincide with the saturation kinetic behavior
for the concentration of HSiMe2Ph in Figure 6: kobsd ) k2k3K1[HSiMe2-
Ph]/(k-2 + k3)(K1+ [PPh3]).
× 10-3, K1 ) 1.1 × 10-4 M, K2 ) 13 M-1
.
We described that the kinetic observations for 1e are
reasonably accounted for by the mechanism in Scheme
7, which is composed of a rapid equilibration between
1e and 9e and the rate-determining conversion of 9e
into the products. It was further confirmed that an
alternative rate expression derived from the steady-
(26) The activation energy (∆Gq) so far reported for the oxidative
addition of hydrosilane toward late-transition-metal complexes falls
within the range 8-15 kcal mol-1 27 However, it is known that, at least
.
for the manganese complexes,27a,d,f the activation energy mainly reflects
the energy change associated with predissociation of a coordinated
solvent such as heptane. Therefore, the actual activation barrier for
oxidative addition may be lower than the observed one.

1316 Organometallics, Vol. 19, No. 7, 2000
Maruyama et al.
structure bearing a highly electron-donating ligand at
the apical position.31 All of the complexes in Figure 7
essentially possess such structural features. Thus, the
most donating alkenyl ligand is located at the apical
position, and the two PPh3 groups and the carbonyl
ligand occupy three of the four corners of the coordina-
tion square, respectively. The only exception is the
orientation of the chloride ligand. Thus, as judged from
the alkenyl(CR)-Ru-Cl angles, its location significantly
deviates from the ideal coordination plane, reflecting the
steric demand of alkenyl ligands. The flexibility of the
Ru-Cl bond may be attributed to its ionic bonding
character.
It is clearly seen that the p-methoxystyryl complex
1f has the space capable of accepting direct interaction
with hydrosilane. It is convincing that the other styryl
complexes (1b, 1g-k ) employed in this study have
similar structural features as well. The approach of
hydrosilane toward ruthenium should take place ini-
tially from the compact and electron-rich hydrogen
atom.16a This process may involve enlargement of the
alkenyl(CR)-Ru-Cl angle, resulting from the flexible
Ru-Cl bond (Scheme 3). The electrophilic nature of the
hydrosilane is enhanced by this coordination, leading
to nucleophilic attack of the R-carbon of the styryl ligand
on silicon (Scheme 5). The four-center transition state
thus formed gives rise to the selective formation of
Ru-H and Si-C bonds (path A, mechanism I in Scheme
4).
In contrast, the space for the interaction with hy-
drosilane in 1d and 1e is effectively blocked by phenyl
(1d ) or silylethenyl (1e) groups at the R-position of
alkenyl ligands. Consequently, these complexes must
dissociate one of the PPh3 ligands prior to the interac-
tion with hydrosilane (Schemes 6 and 7). As supported
by the kinetic data for 1e, the coordinatively unsatur-
ated species thus produced (8d and 8e) undergo oxida-
tive addition of hydrosilane to give hydrido(organo)-
silylruthenium(IV) intermediates, which reductively
eliminate the C-H coupling products (4) along with the
formation of silylruthenium (path B, mechanism IV in
Scheme 8).
F igu r e 7. Chem3D views of the X-ray structures of 1f,
1d , and the bis(trimethylsilyl) analogue of 1e (1e′). Two
PPh3 ligands are omitted for clarity.
for C-H reductive elimination from cis-PtR(H)(PPh3)2
(R ) Me (3.3), CH2CF3 (2.2)).28a,b Hence, mechanism IV
in Scheme 8 has been supported.
Com p a r ison of X-r a y Str u ctu r es. It has been found
that clearly distinct processes are operative for path A
and path B in Scheme 2, respectively. For path A, the
starting alkenylruthenium complex directly reacts with
hydrosilane, without dissociation of PPh3. In contrast,
path B involves prior dissociation of PPh3. Therefore,
one may expect a shift of the reaction course from path
B to path A by addition of PPh3 to the system. The
change in the selectivity observed for 1c (runs 3-6 in
Table 1) should be the case. In contrast, no trace of C-H
coupling products derived from path A were observed
for 1d and 1e even in the presence of free PPh3 (runs 8
and 10). We next examined the reason for this using
X-ray structures.
Figure 7 shows bird’s eye views of 1f, 1d , and the bis-
(trimethylsilyl) analogue of 1e (1e′). The PPh3 ligands
which are located above and below the ruthenium are
omitted for clarity. The structures of 1d and 1e′ are
taken from the literature,23,29 while that of 1f was
confirmed in this study.30
Con clu sion s
We have demonstrated that the C-Si and C-H bond
formation from alkenylruthenium(II) complexes and
hydrosilane exhibit clearly different kinetic behavior
from each other. Although it might be hard to eliminate
explicitly the possibility of the oxidative addition-
reductive elimination process, all of the kinetic data
observed for the C-Si bond formation from styryl
complexes (1b , 1f-k ) (path A) have been reasonably
accounted for by a σ-bond metathesis process, which is
initiated by direct association of the styrylruthenium
with hydrosilane, followed by nucleophilic attack of the
styryl ligand on silicon. This mechanism resembles “the
adduct formation mechanism”, previously proposed for
the silane alcoholysis catalyzed by a cationic iridium-
(III) complex,12a in which nucleophilic attack of a
coordinated alcohol on an η2-hydrosilane ligand was
postulated to be preferred over the conventional oxida-
Organoruthenium(II) complexes with a d6 metal
center have the propensity to make a square-pyramidal
(27) (a) Palmer, B. J .; Hill, R. H. Can. J . Chem. 1996, 74, 1959. (b)
Hostetler, M. J .; Buees, M. D.; Bergman, R. G. Organometallics 1993,
12, 65. (c) Hostetler, M. J .; Bergman, R. G. J . Am. Chem. Soc. 1992,
114, 7629. (d) Hester, D. M.; Sun, J .; Harper, A. W.; Yang, G. K. J .
Am. Chem. Soc. 1992, 114, 5234. (e) Hill, R. H.; Wrighton, M. S.
Organometallics 1985, 4, 413. (f) Hill, R. H.; Wrighton, M. S. Orga-
nometallics 1987, 6, 632.
(28) (a) Abis, L.; Sen, A.; Halpern, J . J . Am. Chem. Soc. 1978, 100,
2915. (b) Michelin, R. A.; Faglia, S.; Uguagliati, P. Inorg. Chem. 1983,
22, 1831. (c) Okrasinski, S. J .; Norton, J . R. J . Am. Chem. Soc. 1977,
99, 295. (d) Keyes, M. C.; Young, V. G., J r.; Tolman, W. B. Organo-
metallics 1996, 15, 4133. (e) Harper, T. G. P.; Desrosiers, P. J .; Flood,
T. C. Organometallics 1990, 9, 2523. (f) Ittel, S. D.; Tolman, C. A.;
English, A. D.; J esson, J . P. J . Am. Chem. Soc. 1978, 103, 3396. (g)
Parkin, G.; Bercaw, J . E. Organometallics 1989, 8, 1172. (h) Buchanan,
J . M.; Stryker, J . M.; Bergman, R. G. J . Am. Chem. Soc. 1986, 108,
1537. (i) J ones, W. D.; Feher, F. J . J . Am. Chem. Soc. 1984, 106, 1650.
(29) (a) Torres, M. R.; Vegas, A.; Santos, A. J . Organomet. Chem.
1986, 309, 169. (b) Torres, M. R.; Vegas, A.; Santos, A.; Ros, J . J .
Organomet. Chem. 1987, 326, 413.
(30) Details of the X-ray structure of 1f are reported in the
Supporting Information.
(31) Albright, T. A.; Burdett, J . K.; Whangbo, M. H. Orbital
Interactions in Chemistry; Wiley: New York, 1985; p 310.

Reactions of Ru Alkenyl Complexes with Hydrosilanes
Organometallics, Vol. 19, No. 7, 2000 1317
(virtual triplet, J ) 5 Hz, PPh), 125.4 (s, C6H4), 113.5 (s, C6H4),
65.8 (s, Et2O), 55.2 (s, MeO), 15.3 (s, Et2O). 31P{1H} NMR
(CDCl3): δ 31.3 (s). IR (KBr): νCO 1923 cm-1. Anal. Calcd for
tive addition process. Common features for both reac-
tions are occurrence of a nucleophilic attack on the
coordinated hydrosilane and negligible kH-Si/kD-Si value.
On the other hand, the C-H bond formation from 1d
and 1e (path B) has been suggested to involve a Ru(IV)
intermediate which is formed by dissociation of PPh3,
followed by oxidative addition of hydrosilane.
C
46H39ClO2P2Ru‚C4H10O: C, 66.00; H, 5.51. Found: C, 65.82;
H, 5.40.
Complexes 1g-k were similarly prepared and isolated as
red crystals by recrystallization from CH2Cl2/Et2O. The spec-
troscopic and analytical data are as follows.
It should be noted that the C-Si and C-H bond
formation selectively produce hydrido- and silylruthe-
nium complexes, respectively, which are the key inter-
mediates for the Chalk-Harrod and modified Chalk-
Harrod cycles. The reaction courses are mainly dictated
by steric conditions around the ruthenium, which
control the reaction processes without or with the
dissociation of PPh3. The former leads to the C-Si bond
formation, while the latter gives rise to the C-H bond
formation.
Ru [CHdCH(C6H4-p-Me)]Cl(CO)(P P h 3)2 (1g). Yield: 74%.
1H NMR (CDCl3): δ 8.35 (dt, J ) 13.0 and 2.1 Hz, 1H, dCH),
7.59-7.53 (m, 12H, Ph), 7.52-7.33 (m, 18H, Ph), 6.96 (d, J )
8.1 Hz, 2H, C6H4), 6.70 (d, J ) 8.1 Hz, 2H, C6H4), 5.55 (dt, J
) 13.0 and 2.1 Hz, 1H, dCH), 2.23 (s, 3H, Me). 13C{1H} NMR
(CDCl3): δ 201.7 (t, J ) 14 Hz, CO), 145.3 (t, J ) 12 Hz,
RuCHdCH), 136.2 (s, C6H4), 135.1 (t, J ) 4 Hz, RuCHdCH),
134.2 (virtual triplet, J ) 6 Hz, PPh), 133.8 (s, C6H4), 131.8
(virtual triplet, J ) 21 Hz, PPh), 130.1 (s, PPh), 128.8 (s, C6H4),
128.3 (virtual triplet, J ) 5 Hz, PPh), 124.4 (s, C6H4), 21.0 (s,
Me). 31P{1H} NMR (CDCl3): δ 31.1 (s). IR (KBr): νCO 1923
cm-1. Anal. Calcd for C46H39ClOP2Ru: C, 68.53; H, 4.88.
Found: C, 68.82; H, 5.02.
Exp er im en ta l Section
Ru [CHdCH(C6H4-p-F )]Cl(CO)(P P h 3)2 (1h ). Yield: 66%.
1H NMR (CDCl3): δ 8.28 (dt, J ) 13.2 and 2.1 Hz, 1H, dCH),
7.58-7.52 (m, 12H, Ph), 7.44-7.34 (m, 18H, Ph), 6.81 (t, J )
8.8 Hz, 2H, C6H4), 6.69 (t, J ) 8.8 Hz, 2H, C6H4). 13C{1H} NMR
(CDCl3): δ 201.6 (t, J ) 16 Hz, CO), 160.4 (d, 1J C-F ) 243 Hz,
C6H4), 145.9 (br, RuCHdCH), 135.1 (s, RuCHdCH), 134.2
(virtual triplet, J ) 6 Hz, PPh), 131.7 (virtual triplet, J ) 22
Hz, PPh), 130.2 (s, PPh), 128.3 (virtual triplet, J ) 5 Hz, PPh),
125.4 (d, J C-F ) 8 Hz, C6H4), 114.7 (d, J C-F ) 21 Hz, C6H4).
One carbon signal of the C6H4F ring could not be detected,
probably due to overlapping. 31P{1H} NMR (CDCl3): δ 31.4
(s). IR (KBr): νCO 1912 cm-1. Anal. Calcd for C45H36ClFOP2-
Ru: C, 66.71; H, 4.48. Found: C, 66.50; H, 4.47.
Ru [CHdCH(C6H4-p-Cl)]Cl(CO)(P P h 3)2‚Et2O (1i‚Et2O).
Yield: 57%. 1H NMR (CDCl3): δ 8.44 (dt, J ) 13.5 and 2.1
Hz, 1H, dCH), 7.57-7.51 (m, 12H, Ph), 7.45-7.33 (m, 18H,
Ph), 7.08 (d, J ) 8.4 Hz, 2H, C6H4), 6.66 (d, J ) 8.4 Hz, 2H,
C6H4), 5.50 (dt, J ) 13.5 and 2.1 Hz, 1H, dCH), 5.51 (dt, J )
13.0 and 2.1 Hz, 1H, dCH), 3.47 (q, J ) 7.0 Hz, 4H, Et2O),
1.20 (t, J ) 7.0 Hz, 6H, Et2O). 13C{1H} NMR (CDCl3): δ 201.4
(t, J ) 15 Hz, CO), 148.3 (t, J ) 11 Hz, RuCHdCH), 137.0 (t,
J ) 2 Hz, RuCHdCH), 134.1 (virtual triplet, J ) 6 Hz, PPh),
131.6 (virtual triplet, J ) 22 Hz, PPh), 130.2 (s, PPh, para),
129.7 (s, C6H4), 128.3 (virtual triplet, J ) 5 Hz, PPh), 128.2
(s, C6H4), 125.4 (s, C6H4), 65.8 (s, Et2O), 15.3 (s, Et2O). 31P-
{1H} NMR (CDCl3): δ 31.6 (s). IR (KBr): νCO 1925 cm-1. Anal.
Calcd for C45H36Cl2OP2Ru‚C4H10O: C, 65.33; H, 5.15. Found:
C, 65.08; H, 4.92.
Gen er a l Con sid er a tion s. All manipulations were carried
out under a nitrogen atmosphere using standard Schlenk
techniques. Nitrogen gas was dried by passage through P2O5
(Merck, SICAPENT). NMR spectra were recorded on a Varian
Mercury 300 (1H NMR, 300.106 MHz; 13C NMR, 75.470 MHz;
31P NMR, 121.486 MHz) spectrometer. Chemical shifts are
reported in δ (ppm) referred to an internal SiMe4 standard
for 1H and 13C NMR and to an external 85% H3PO4 standard
for 31P NMR. Mass spectra were measured using a Shimadzu
QP-5000 GC-mass spectrometer (EI, 70 eV, capillary column).
GLC analysis was performed with a GL Sciences GC-353
instrument equipped with a FID detector and a capillary
column (TC-1, 30 m). THF, Et2O, benzene, and hexane were
dried over sodium benzophenone ketyl and distilled just before
using. CH2Cl2 was dried over CaH2 and distilled just before
using. CDCl3 was purified by passage though an Al2O3 column
and dried over molecular sieves 4A. The complexes RuHCl-
(CO)(PPh3)3 (2),32 Ru(CHdCHPh)Cl(CO)(PPh3)2 (1b ),29a
Ru[CHdCH(t-Bu)]Cl(CO)(PPh3)2 (1c),29b Ru(CPhdCHPh)Cl-
(CO)(PPh3)2 (1d ),29a and Ru[C(dCHSiMe2Ph)CHdCHSiMe3]Cl-
(CO)(PPh3)2 (1e)8 were synthesized according to the literature.
All other compounds were obtained from commercial sources
and used without purification.
3
2
P r ep a r a tion of P a r a -Su bstitu ted Styr yl Com p lexes
(1f-k ). The preparation of Ru[CHdCH(C6H4-p-OMe)]Cl(CO)-
(PPh3)2 (1f) represents a typical example. The complex RuHCl-
(CO)(PPh3)3 (2; 201 mg, 0.211 mmol) was dissolved in CH2Cl2
(10 mL) at room temperature, and (p-methoxyphenyl)acetylene
(40 µL, 0.31 mmol) was added by means of a syringe. The
yellow solution instantly turned red. After 15 min the solution
was filtered through a short Florisil column and concentrated
under reduced pressure. Addition of hexane (3 mL) with
stirring led to precipitation of an orange solid, which was
collected by filtration, washed with hexane (2 mL × 2), and
dried under vacuum. The crude product was dissolved in CH2-
Cl2 (1 mL), and Et2O (3 mL) was carefully layered. The solvent
layers were mixed slowly at -20 °C to give red crystals of 1f,
which contains 1 equiv of Et2O in the crystal (127 mg, 73%
Ru [CHdCH(C6H4-p-Br )]Cl(CO)(P P h 3)2‚Et2O (1j‚Et2O).
Yield: 69%. 1H NMR (CDCl3): δ 8.50 (dt, J ) 13.4 and 2.0
Hz, 1H, dCH), 7.57-7.51 (m, 12H, Ph), 7.45-7.31 (m, 18H,
Ph), 7.23 (d, J ) 8.4 Hz, 2H, C6H4), 6.62 (d, J ) 8.4 Hz, 2H,
C6H4), 5.49 (dt, J ) 13.4 and 2.0 Hz, 1H, dCH), 3.47 (q, J )
7.0 Hz, 4H, Et2O), 1.20 (t, J ) 7.0 Hz, 6H, Et2O). 13C{1H} NMR
(CDCl3): δ 201.5 (t, J ) 14 Hz, CO), 148.9 (t, J ) 11 Hz,
RuCHdCH), 137.4 (s, RuCHdCH), 134.1 (virtual triplet, J )
6 Hz, PPh), 131.6 (virtual triplet, J ) 22 Hz, PPh), 131.1 (s,
C6H4), 130.2 (s, PPh), 128.3 (virtual triplet, J ) 5 Hz, PPh),
125.8 (s, C6H4), 117.7 (s, C6H4), 65.8 (s, Et2O), 15.3 (s, Et2O).
31P{1H} NMR (CDCl3): δ 31.4 (s). IR (KBr): νCO 1925 cm-1
.
1
yield). H NMR (CDCl3): δ 8.13 (dt, J ) 13.2 and 2.2 Hz, 1H,
Anal. Calcd for C45H36BrClOP2Ru‚C4H10O: C, 62.26; H, 4.91.
Found: C, 61.58; H, 4.68.
dCH), 7.60-7.53 (m, 12H, Ph), 7.43-7.33 (m, 18H, Ph), 6.70
(s, 4H, C6H4), 5.50 (dt, J ) 13.2 and 2.2 Hz, 1H, dCH), 3.76
(s, 3H, OCH3), 3.47 (q, J ) 7.0 Hz, 4H, Et2O), 1.20 (t, J ) 7.0
Hz, 6H, Et2O). 13C{1H} NMR (CDCl3): δ 201.7 (t, J ) 14 Hz,
CO), 156.9 (s, C6H4), 143.0 (br, RuCHdCH), 134.7 (s, RuCHd
CH), 134.1 (virtual triplet, J ) 6 Hz, PPh), 132.2 (s, C6H4),
131.8 (virtual triplet, J ) 21 Hz, PPh), 130.1 (s, PPh), 128.3
Ru [CHdCH(C6H4-p-CF3)]Cl(CO)(P P h 3)2‚Et2O (1k‚Et2O).
Yield: 66%. 1H NMR (CDCl3): δ 8.76 (dt, J ) 13.4 and 2.0
Hz, 1H, dCH), 7.57-7.50 (m, 12H, Ph), 7.45-7.33 (m, 20H,
Ph and C6H4), 6.81 (d, J ) 8.2 Hz, 2H, C6H4), 5.60 (dt, J )
13.4 and 2.0 Hz, 1H, dCH), 3.47 (q, J ) 7.0 Hz, 4H, Et2O),
1.20 (t, J ) 7.0 Hz, 6H, Et2O). 13C{1H} NMR (CDCl3): δ 201.3
(t, J ) 14 Hz, CO), 152.6 (br, RuCHdCH), 141.4 (s, C6H4),
134.3 (s, RuCHdCH), 134.1 (virtual triplet, J ) 6 Hz, PPh),
(32) Ahmad, N.; Levison, J . J .; Robinson, S. D.; Uttley, M. F. Inorg.
Synth. 1974, 15, 45.

1318 Organometallics, Vol. 19, No. 7, 2000
Maruyama et al.
131.5 (virtual triplet, J ) 22 Hz, PPh), 130.3 (s, PPh), 128.4
(virtual triplet, J ) 5 Hz, PPh), 125.8 (q, 2J C-F ) 32 Hz, C6H4),
125.1 (q, J C-F ) 4 Hz, C6H4), 124.6 (q, J C-F ) 272 Hz, CF3),
124.1 (s, C6H4), 65.8 (s, Et2O), 15.3 (s, Et2O). 31P{1H} NMR
(CDCl3): δ 31.6 (s). IR (KBr): νCO 1926 cm-1. Anal. Calcd for
7.32 (m, 6H, Ar), 6.88 (d, J ) 18.9 Hz, 1H, dCHAr), 6.41 (d, J
) 18.9 Hz, 1H, dCHSi), 3.81 (s, 3H, MeO), 0.42 (s, 6H, SiMe2).
MS (m/z (relative intensity, %)): 268 (M+, 40), 253 (53), 176
(17), 175 (100), 145 (39), 121 (33), 105 (17), 59 (21), 43 (56).
Identification data for the other hydrosilylation products are
as follows.
3
1
C
46H36F3ClOP2Ru‚C4H10O: C, 64.27; H, 4.96. Found: C, 64.51;
H, 4.94.
Reaction s of Alken yl Com plexes with HSiMe2P h (Table
tr a n s-(p-MeC6H4)CHdCHSiMe2P h (4g). 1H NMR
(CDCl3): δ 7.62-7.33 (m, 7H, Ar), 7.13 (d, J ) 8.1 Hz, 2H,
Ar), 6.91 (d, J ) 18.9 Hz, 1H, dCHAr), 6.51 (d, J ) 18.9 Hz,
1H, dCHSi), 2.34 (s, 3H, Me), 0.41 (s, 6H, SiMe2). MS (m/z
(relative intensity, %)): 252 (M+, 37), 237 (58), 160 (29), 159
(100), 145 (41), 135 (17), 121 (35), 118 (17), 105 (19), 43 (56).
1). A typical procedure (run 2) is as follows. Complex 1b (4.8
mg, 6.0 µmol) and PPh3 (1.6 mg, 6.0 mmol) were loaded in an
NMR sample tube equipped with a rubber septum cap, and
the system was replaced with nitrogen gas. Anisole (1.0 µL,
9.2 µmol) was added as an internal standard for 1H NMR
analysis. All contents were dissolved in CDCl3 (0.6 mL) at room
temperature and then cooled to -40 °C. HSiMe2Ph (16.4 mg,
0.120 mmol) was added, and the sample solution was warmed
to 30 °C. The color of the solution instantly changed from red
to orange. 31P{1H} NMR analysis revealed the absence of 1b
and the formation of the hydride complex 2 in over 99%
selectivity. The amounts of trans-PhCHdCHSiMe2Ph (4b)33
and styrene (5b) produced were determined on the basis of
the relative peak integration of olefinic protons of 4b (δ 6.78
(d)) and 5b (δ 5.73 (d)) to anisole (δ 3.81 (s)) in the 1H NMR
spectrum. The formation of 4b and 5b was confirmed also by
GC-MS spectrometry.
tr a n s-(p-F C6H4)CHdCHSiMe2P h
(4h ).
1H
NMR
(CDCl3): δ 7.67-7.31 (m, 7H, Ar), 7.02 (t, J ) 8.7 Hz, 2H,
Ar), 6.90 (d, J ) 18.9 Hz, 1H, dCHAr), 6.50 (d, J ) 18.9 Hz,
1H, dCHSi), 0.44 (s, 6H, SiMe2). MS (m/z (relative intensity,
%)): 256 (M+, 27), 241 (54), 164 (15), 163 (100), 145 (27), 121
(52), 105 (18), 53 (15), 47 (30), 43 (64).
tr a n s-(p-ClC6H4)CHdCHSiMe2P h
(4i).
1H
NMR
(CDCl3): δ 7.64-7.53 (m, 3H, Ar), 7.46-7.35 (m, 6H, Ar), 6.87
(d, J ) 18.9 Hz, 1H, dCHAr), 6.56 (d, J ) 18.9 Hz, 1H, d
CHSi), 0.42 (s, 6H, SiMe2). MS (m/z (relative intensity, %)):
274 (M + 2, 9), 272 (M+, 26), 257 (46), 181 (28), 179 (100), 145
(28), 121 (52), 105 (21), 63 (21), 43 (74).
tr a n s-(p-Br C6H4)CHdCHSiMe2P h
(4j). 1H
NMR
(CDCl3): δ 7.62-7.29 (m, 9H, Ar), 6.85 (d, J ) 18.9 Hz, 1H,
dCHAr), 6.57 (d, J ) 18.9 Hz, 1H, dCHSi), 0.43 (s, 6H, SiMe2).
MS (m/z (relative intensity, %)): 318 (M + 2, 1), 316 (M+, 1),
225 (3), 223 (3), 179 (4), 121 (4), 44 (100), 43 (9), 41 (3), 40
(34).
The reactions of 1c-e with HSiMe2Ph were similarly
examined. The ratio of RuHCl(CO)(PPh3)3 (2; δ 39.6 (br) and
13.5 (br)) to Ru(SiMe2Ph)Cl(CO)(PPh3)2 (3;8,34 δ 33.4 (s)) was
determined by 31P{1H} NMR spectroscopy. Organic products
1
were analyzed by H NMR spectroscopy and GC-MS spectrom-
tr a n s-(p-CF 3C6H4)CHdCHSiMe2P h (4k ). 1H NMR
(CDCl3): δ 7.62-7.28 (m, 9H, Ar), 6.95 (d, J ) 19.2 Hz, 1H,
dCHAr), 6.70 (d, J ) 19.2 Hz, 1H, dCHSi), 0.45 (s, 6H, SiMe2).
MS (m/z (relative intensity, %)): 306 (M+, 34), 291 (52), 229
(34), 213 (90), 151 (72), 145 (29), 121 (88), 105 (29), 77 (27), 43
(100).
etry. All are known compounds: t-BuCHdCHSiMe2Ph (4c),33
(1E,3E)-1-(dimethylphenylsilyl)-4-(trimethylsilyl)-1,3-butadi-
ene (5e′),8 and 1-(dimethylphenylsilyl)-4-(trimethylsilyl)-2-
butene (5e′′).8
Kin etic Stu d y on th e Rea ction of 1b w ith HSiMe2P h
(Ta ble 2). A typical procedure (run 4) is as follows. Complex
1b (4.8 mg, 6.0 µmol) and PPh3 (1.6 mg, 6.0 µmol) were loaded
in an NMR sample tube equipped with a rubber septum cap,
and the system was replaced with nitrogen gas. Anisole (1 µL,
9.2 µmol) was added. All contents were dissolved in CDCl3 (ca.
0.5 mL) at room temperature. HSiMe2Ph (16.4 mg, 0.120
mmol) was added at -40 °C, and the volume of the solution
was adjusted with additional CDCl3 exactly to 0.60 mL. The
sample was placed in an NMR sample probe controlled to 10.0
( 0.1 °C. The amounts of 1b at intervals were determined by
measuring the relative peak integration of the olefinic proton
signal of 1b at δ 5.58 and the methyl proton signal of anisole
at δ 3.81 in 1H NMR spectra.
Kin et ic St u d ies on t h e R ea ct ion of 1d a n d 1e w it h
HSiMe2P h . A typical procedure is as follows. Complex 1d (5.3
mg, 6.0 µmol) and PPh3 (1.6 mg, 6.0 µmol) were loaded in an
NMR sample tube, and the system was replaced with nitrogen
gas. Anisole (0.6 µL, 5.5 µL) and HSiMe2Ph (16.4 mg, 0.120
mmol) were added at 0 °C by means of syringe. The contents
were dissolved in CDCl3 (ca. 0.5 mL), and the volume of the
solution was adjusted to 0.60 mL with additional CDCl3 at
room temperature. The sample tube was placed in an NMR
sample probe controlled to 45.0 ( 0.1 °C, and the time course
of the reaction was followed by 1H NMR spectroscopy. The
amounts of 1d at intervals were determined on the basis of
the relative peak integration of the olefinic proton signal of
1d at δ 5.43 and the methyl proton signal of anisole at δ 3.81.
Kinetic experiments on the reaction of 1e were similarly
conducted. The time course was followed by measuring the
olefinic proton signal of 1e at δ 5.84.
Kinetic runs with the series of para-substituted styryl
complexes 1f-k were similarly conducted. The product styr-
ylsilanes were identified by 1H NMR spectroscopy and GLC
analysis using authentic samples, which were prepared by
catalytic hydrosilylation of phenylacetylene derivatives.
Ca ta lytic Hyd r osilyla tion of P a r a -Su bstitu ted P h e-
n yla cetylen es w ith HSiMe2P h . A typical procedure is as
follows. To a Schlenk tube containing a catalytic amount of
RuHCl(CO)(PPh3)3 (50.0 mg, 0.0525 mmol) were added CH2-
Cl2 (1 mL), (p-methoxyphenyl)acetylene (137 mg, 1.05 mmol),
and HSiMe2Ph (157 mg, 1.15 mmol) at room temperature. The
mixture was stirred at 30 °C for 3 h. GLC analysis revealed
the absence of the starting acetylene and the selective forma-
tion of trans-(p-MeOC6H4)CHdCHSiMe2Ph (4f), which was
isolated by column chromatography (SiO2, CH2Cl2) followed
by bulb-to-bulb distillation under reduced pressure (258 mg,
Ack n ow led gm en t. This work was supported by a
Grant-in-Aid for Scientific Research on Priority Area
“The Chemistry of Inter-element Linkage” (No. 09239105)
from the Ministry of Education, Science, Sports and
Culture of J apan.
Su p p or tin g In for m a tion Ava ila ble: Details of the X-ray
structural determination of 1f‚Et2O, including an experimental
description, an ORTEP drawing, an atomic numbering scheme,
and tables of crystal data, atomic coordinates, thermal pa-
rameters, and all bond distances and angles, and tables giving
kinetic data for 1d and 1e. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
92% yield). H NMR (CDCl3): δ 7.59-7.52 (m, 3H, Ar), 7.41-
(33) J un, C.-H.; Crabtree, R. H. J . Organomet. Chem. 1993, 447,
177.
(34) Marciniec, B.; Pietraszuk, C. Organometallics 1997, 16, 4320.
OM9909035