Ruthenium-catalyzed hydrosilylation of 1-alkynes with novel regioselectivity

Article abstract of DOI:10.1021/ol026089q

(figure presented) A ruthenium catalyst precursor bearing a bulky and electron-donating pentamethylcyclopentadienyl (Cp*) ligand, typically Cp*RuH₃(PPh₃), mediates hydrosilylation of several 1-alkynes with novel regioselectivity to g

Full text of DOI:10.1021/ol026089q

ORGANIC
LETTERS
2002
Vol. 4, No. 17
2825-2827
Ruthenium-Catalyzed Hydrosilylation of
1-Alkynes with Novel Regioselectivity
Yukio Kawanami,* Yoshiya Sonoda, Takashi Mori, and Keiji Yamamoto*
Department of Materials Science and EnVironmental Engineering, Tokyo UniVersity of
Science, Yamaguchi, 1-1-1 Daigaku-Dori, Onoda 756-0884, Japan
yamamoto@ed.yama.tus.ac.jp
Received April 27, 2002
ABSTRACT
A ruthenium catalyst precursor bearing a bulky and electron-donating pentamethylcyclopentadienyl (Cp*) ligand, typically Cp*RuH (PPh3),
3
mediates hydrosilylation of several 1-alkynes with novel regioselectivity to give preferentially 2-silyl-1-alkenes.
Hydrosilylation of 1-alkynes is a versatile means for prepar-
ing 1-alkenylsilanes, which are widely used for organic
synthesis.1 Thus, regio- and stereocontrol of the reaction is
indispensable for the synthetic use of 1-alkenylsilanes.
Hydrosilylation of 1-alkynes catalyzed by various transition
metal complexes such as chloroplatinic acid often gives a
mixture of three possible isomers (eq 1).
Takeuchi2 and Ojima,3 suppressing the formation of an
internal adduct in both cases. On the other hand, regiocontrol
to yield an internal adduct has received less attention.4
However, recent reports do describe the selective formation
of the internal adduct in the ruthenium complex-catalyzed
hydrosilylation of alkynes, either by the necessary functional
group-directed addition of trialkyl- and trialkoxysilanes5 or
by a more general Markovnikov-type one.6 We have already
found that the complex Cp*RuH3(PPh3) (1)7 is a unique
catalyst precursor for the regiocontrolled hydrosilylation of
1-alkynes with preferably dichloromethylsilane to afford
mainly 2-silyl-1-alkenes8 and report herein our own results.
(2) (a) Takeuchi, R.; Ebata, I. Organometallics 1997, 16, 3707-3710.
(b) Takeuchi, R.; Nitta, S. J. Org. Chem. 1995, 60, 3045-3051. (c)
Takeuchi, R.; Tanouchi, N. J. Chem. Soc., Perkin Trans. 1 1994, 2909-
2913. (d) Takeuchi, R.; Nitta, S.; Watanabe, D. J. Chem. Soc., Chem.
Commun. 1994, 1777-1778. (e) Takeuchi, R.; Tanouchi, N. J. Chem. Soc.,
Chem. Commun. 1993, 1319-1320.
Both (E)- and (Z)-1-silyl-1-alkenes are obtained via the
addition of a hydrosilane across the carbon-carbon triple
bond, where the silicon atom attaches to the terminal carbon,
being referred to as terminal adducts. The reversed addition
of a hydrosilane to 1-alkynes gives a 2-silyl-1-alkene as an
internal adduct. Rigorous stereocontrol to yield either (E)-
or (Z)-1-silyl-1-alkene has been achieved independently by
(3) Ojima, I.; Clos, N.; Donovan, R. J.; Ingallina, P. Organometallics
1990, 9, 3127-3133.
(4) Using rather uncommon alkynes: Matsumoto, H.; Hoshino, Y.; Nagai,
Y. Chem. Lett. 1982, 1663-1666.
(5) Hydroxyl group-directed reaction: Na, Y.; Chang, S. Org. Lett. 2000,
2, 1887-1889. Phenylacetylene as a substrate: Voronkov, M. G.; Pukhnarev-
ich, V. B.; Tsykhanskaya, I. I.; Ushakova, N. I. Inorg. Chim. Acta 1983,
68, 103-105.
(6) Trost, B. M.; Ball, Z. T. J. Am. Chem. Soc. 2001, 123, 12726-12727.
(7) Suzuki, H.; Lee, D. H.; Oshima, N.; Moro-oka, Y. Organometallics
1987, 6, 1569-1575.
(8) Kawanami, Y.; Yamamoto, K. 74th National Meeting of Chem. Soc.
Jpn., Kyoto, March, 1998; Abstract 4D828.
(1) (a) Fleming, I.; Dunogues, J.; Smithers, R. H. Org. React. 1989, 37,
57-575. (b) Colvin, E. W. Silicon Reagents in Organic Synthesis; Academic
Press: London, 1988. (c) Weber, W. P. Silicon Reagents for Organic
Synthesis; Springer-Verlag: Berlin, 1983. (d) Colvin, E. W. Silicon in
Organic Synthesis; Butterworth: London, 1981.
10.1021/ol026089q CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/27/2002

A trihydride complex 1 was employed as a catalyst
precursor. The reaction of 1-heptyne (2 mmol) and HSiCl3
(3 mmol) in toluene with a catalytic amount of 1 (0.06 mmol,
3 mol % to 1-heptyne) at 80 °C for 24 h (Table 1, entry 1)
that, in most reactions that gave internal adducts (2)
preferentially, the minor terminal adduct (3) mainly has the
(Z)-configuration, which must arise from an overall anti
addition of hydrosilane to 1-alkyne.
When the reactions of 1-heptyne were carried out with
trialkylsilanes instead of chlorosilanes, simple dimerization
of 1-heptyne took place along with the hydrosilylation of
1-heptyne in each case. For example, reaction of 1-heptyne
and Et2MeSiH, under the otherwise same conditions as
described above, was sluggish and gave both a regioisomeric
mixture of dimers10 (48% yield) and adducts of hydrosilane
(34% yield), the latter consisting mainly of an internal adduct
(I:T ) 82:18). This kind of dimer formation has already been
reported for ruthenium-catalyzed reactions in the absence of
hydrosilane.11,12
Table 1. Ruthenium-Catalyzed Hydrosilylation of Terminal
Alkynes with Chlorosilanesa
The effect of various ruthenium complexes on the catalytic
activity for the hydrosilylation of 1-heptyne with HSiMeCl2
was examined (Table 2). Reactions catalyzed by Cp*Ru(II)
entry
alkyne R
silane X3
I:T (E:Z)b
yield, % (2 + 3)c
1
2
3
4
5
6
7
8d
9
n-C5H11
Cl3
69:31 (1:1)
89:11 (1:7)
85:15 (1:9)
92:8 (0:1)
93:7 (0:1)
97:3 (0:1)
>99:1
83:17 (1:1)
69:31 (1:1)
17:83 (1:0)
71:29 (1:0)
79
89
65
89
68
83
78
71
81
82
18
MeCl2
Me2Cl
MeCl2
MeCl2
MeCl2
MeCl2
MeCl2
MeCl2
MeCl2
MeCl2
Table 2. Hydrosilylation of 1-Heptyne with
Dichloromethylsilane Using Various Ruthenium Catalysts
n-C6H13
AcO(CH2)3
AcO(CH2)2
PhOCH2
i-AmOCH2
Ph
entry
catalyst
I:T (E:Z)a
yield, % (2 + 3)b
1
2
3
4
5
Cp*RuH3(PPh3)
Cp*RuH3(PCy3)
CpRuH(PPh3)2
CpRuCl(PPh3)2
Ru(OAc)2(PPh3)
89:11 (1:7)
93:7 (0:1)
30:70 (1:2)
39:61 (1:1)
25:75 (1:4)
89
93
29c
7d
10
11
Me3C
Me3Si
49
a Reaction conditions: alkyne (2 mmol) and silane (3 mmol) in toluene
(4 mL) at 80 °C for 24 h under an argon atmosphere. Cp*RuH3(PPh3) (0.06
mmol) was a catalyst unless otherwise noted. b Determined by GLC and
1H NMR. c Isolated yield. d i-Am ) (CH3)2CH(CH2)2.
a Determined by GLC and 1H NMR measurement. b Isolated yield.
c Reaction time: 7 days. d Reaction time: 3 days.
species exhibited high selectivities for the internal adduct
(entries 1 and 2), while the catalyst precursors without a Cp*
ligand were less effective in terms of both reactivity and
regioselectivity. The bulkier and more electron-donating
phosphorus ligand PCy3 (Cy ) cyclohexyl) was found to
enhance the regioselectivity (entry 2, I:T ) 93:7) in
comparison with PPh3.
Salient features of the present Cp*Ru(II)-catalyzed hy-
drosilylation of 1-alkynes are (i) the anomalous regioselec-
tivity of preferentially giving 2-silyl-1-alkenes (internal
adducts I) rather than 1-silyl-1-alkenes (terminal adducts T),
(ii) formal anti addition that selectively gives the (Z)-isomer
of T, and (iii) that the ruthenium catalyst bearing a Cp*
ligand is essential for the reaction.
On the basis of these features and an NMR observation
of the stoichiometric mixture of 1 and HSiMeCl2 for
liberation of hydrogen, a plausible catalytic loop of the
present hydrosilylation of 1-alkyne is depicted as Figure 1,
where the key intermediate A would originate from the
precursor 1, the hydrosilane, and the substrate alkyne. Thus,
gave an internal adduct I, 2-(trichlorosilyl)-1-heptene, and
a terminal adduct T, 1-(trichlorosilyl)-1-heptene, in a ratio
of 69:31 by GLC analysis. Treatment of the reaction mixture
with EtOH and Et3N in CH2Cl2 gave the corresponding
mixture of 2-(triethoxylsilyl)-1-heptene (2) and 1-(triethoxy-
silyl)-1-heptene (3) in 79% yield by bulb-to-bulb distillation.
Higher regioselectivity in obtaining the 2-silylated adduct
was attained when other chlorosilanes, HSiMeCl2 (entry 2,
I:T ) 89:11) and HSiMe2Cl (entry 3, 85:15), were used
instead of HSiCl3. Reactions of several terminal alkynes with
HSiMeCl2 were examined: both 1-octyne and functionalized
4-pentynyl acetate also gave the internal adduct with slightly
higher selectivity (entries 4 and 5). Excellent selectivity and
good yield were achieved with 3-butynyl acetate and phenyl
propargyl ether (entries 6 and 7, 97:3 and >99:1, respec-
tively), while the reaction of 3-methylbutyl propargyl ether
resulted in lower selectivity with moderate yield (entry 8).
The reaction of phenylacetylene proceeded with inferior
selectivity, still exhibiting the same tendency (entry 9,
69:31). However, the reaction of tert-butylacetylene did
exhibit inverted selectivity (entry 10, 17:83) in contrast to
that of trimethylsilylacetylene, which occurred only in low
yield (entry 11, 71:29, 18% yield9). It should also be noted
(10) 1H NMR signals of the vinylic protons of the dimers obtained were
consistent with the reported spectral data for ruthenium-catalyzed dimer-
ization of 1-hexyne. See Yi’s (ref 11) and Kirchner’s (ref 12) reports.
(11) (a) Yi, C. S.; Liu, N. Organometallics 1996, 15, 3968-3971. (b)
Yi, C. S.; Liu, N. Synlett 1999, 281-287.
(9) Most of trimethylsilylacetylene remained unchanged after 24 h. The
adducts I and T decomposed after a prolonged reaction time (up to 72 h).
(12) Slugovc, C.; Mereiter, K.; Zobetz, E.; Schmid, R.; Kirchner, K.
Organometallics 1996, 15, 5275-5277.
2826
Org. Lett., Vol. 4, No. 17, 2002

consistent with the fact that the ruthenium catalysts bearing
no Cp* ligand exhibited an inferior and even opposite
regioselectivity in the products, probably due to the dimin-
ished steric requirements around the ruthenium center in both
B and B′.
The fact that the minor terminal adduct T forms mainly
with the (Z)-configuration may well be understood as follows.
An existing steric repulsion between the ruthenium center
and a silyl substituent across the carbon-carbon double bond
in B′ would be released once B′ undergoes geometrical
isomerization into B′′ by way of a possible metal carbene
zwitterionic complex, which has been proposed to explain
an apparent anti addition of a hydrosilane to phenylacetylene
catalyzed by rhodium3 and iridium complexes.15 Again,
formal σ-bond metathesis between the hydrosilane and B′′
would result in a terminal adduct T with a preferential (Z)-
configuration and regenerate the key intermediate A. Such
an isomerization in B as well as in B′ might take place,
resulting in the internal adduct I, which is indistinguishable
from that formed directly from B. To test this possibility,
the reaction of 1-heptyne-d1 with HSiMeCl2 using the
complex 1 as an catalyst was examined to give mainly the
internal adduct I, an NOE measurement of which is indicative
of the formal anti addition of the hydrosilane across the triple
bond.16
Figure 1. Plausible silylruthenation pathway for the hydrosilylation
of 1-alkynes.
a key step may involve the silylruthenation, probably in a
manner syn to the coordinated alkyne to form either
â-silylated alkenylruthenium B or B′. This step is akin to
that of dehydrogenative silylation of alkynes mediated by
iridium complexes13 but different from the conventional
catalytic pathway proposed by Chalk and Harrod.14 The
intermediate B would be sterically less demanding and
constitutes the major loop, while B′ may be more congested
due to an additional steric repulsion between the ruthenium
center and the substituent R, becoming the minor one.
Subsequently, the presence of a bulky Cp* ligand and, at
the same time, bulkier PCy3 than PPh3 as a supporting ligand
on the ruthenium center in A would help silylruthenation
proceed highly regioselectively in favor of B rather than B′.
Thus, formal σ-bond metathesis between the hydrosilane and
B would follow to preferentially give the internal adduct I
and regenerate the key intermediate A. These arguments are
In conclusion, we have shown that Cp*Ru(II) complex-
catalyzed hydrosilylation of 1-alkynes proceeds with novel
regioselectivity to afford predominantly internal adducts I.
Acknowledgment. Financial support by the “Grant-in-
Aid for Encouragement of Young Scientists” (11750753)
from Japan Society for the Promotion of Science is gratefully
acknowledged.
Supporting Information Available: Representative pro-
cedure for the hydrosilylation and spectral data of products.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL026089Q
(15) This sort of geometrical isomerization has also been suggested by
Crabtree for the iridium-catalyzed hydrosilylation via the η2-vinylmetal
tautomers (ref. 13), the so-called Ojima-Crabtree mechanism.
(16) By an NOE effect (2%) observed between allylic protons and one
of vinylidene protons (cis to each other), another vinylidene proton in the
ordinary internal adduct 2 obtained from 1-heptyne could easily be assigned,
while the product from 1-heptyne-d1 was found to lack the latter vinylidene
proton, indicating clearly (Z)-1-deuterio-2-(diethoxymethylsilyl)-1-heptene.
(13) Jun, C.-H.; Crabtree, R. H. J. Organomet. Chem. 1993, 447, 177-
187. Tanke, R. S.; Crabtree, R. H. J. Am. Chem. Soc. 1990, 112, 7984-
7989.
(14) Chalk, A. J.; Harrod, J. F. J. Am. Chem. Soc. 1965, 87, 16.
Org. Lett., Vol. 4, No. 17, 2002
2827