Regio- and stereoselective hydrosilylation of terminal alkynes using Grubbs' first-generation olefin-metathesis catalyst

Article abstract of DOI:10.1039/b409832c

Grubbs' first-generation, Ru metathesis complex 3 catalyses the hydrosilylation of terminal alkynes. The reaction exhibits an interesting selectivity profile that is dependent on the reaction concentration and more importantly on the silane employed.

Full text of DOI:10.1039/b409832c

View Article Online / Journal Homepage / Table of Contents for this issue
C O M M U N I C A T I O N
Regio- and stereoselective hydrosilylation of terminal alkynes
using Grubbs’ first-generation olefin-metathesis catalyst†
Caterina S. Aricó and Liam R. Cox*
School of Chemistry, The University of Birmingham, Edgbaston, Birmingham, UK B15 2TT.
E-mail: l.r.cox@bham.ac.uk; Fax: +44 (0)121 414 4403; Tel: +44 (0)121 414 3524
Received 30th June 2004, Accepted 30th July 2004
First published as an Advance Article on the web 13th August 2004
Grubbs’ first-generation, Ru metathesis complex 3 catalyses the
hydrosilylation of terminal alkynes. The reaction exhibits an
interesting selectivity profile that is dependent on the reaction
concentration and more importantly on the silane employed.
Introduction
Vinylsilanes have emerged as powerful intermediates in organic
synthesis, and further advances continue to enhance their synthetic
utility.1 Most notably, they now make viable substitutes for highly
toxicvinylstannanesinPd-catalysedcross-couplingreactions.2 They
have also been successfully employed in ring-closing-metathesis3
and cross-metathesis4 reactions. However, the synthetic utility of
this functionality remains dependent on being able to prepare the
Scheme 2 Selective hydrosilylation for the -2 regioisomer is possible
starting vinylsilane with high levels of regio- and stereocontrol.
Although vinylsilanes can be prepared in a number of ways,5 the
transition-metal-catalysed hydrosilylation of alkynes has proven to
be the most powerful.6 Hydrosilylation of a terminal alkyne using
a silane, R3SiH, in the presence of a transition-metal catalyst can
provide up to three, isomeric vinylsilane products, -(E)-1, -(Z)-1,
and -2 (Scheme 1).
in some cases.
numerous synthetic applications.10 Whilst metathesis remains the
most important application for this class of Ru alkylidene complex,
a growing number of other processes are also mediated by Grubbs’
catalyst.11 For example, Lee and co-workers recently reported that
3 displays excellent catalytic activity in the hydrosilylation of
carbonyl compounds and dehydrogenative condensation of alcohols
and silanes (Scheme 3).12
Scheme 1 Hydrosilylation of terminal alkynes provides three isomeric
vinylsilane products.
The regio- and stereochemical outcome of this reaction depends
on several factors including the choice of catalyst (the metal
and ligands), the substituents on both reacting partners, and to
a lesser extent, on the reaction conditions employed (reaction
temperature, solvent and catalyst loading). The -(E)-stereoisomer,
-(E)-1, is best obtained using Pt-based complexes,7 whilst the
selective formation of the stereoisomeric -(Z)-vinylsilane,
-(Z)-1, is generally achieved using Rh- and Ru-based catalysts.8,9
The formation of the third regioisomer, -2, is best accomplished by
incorporatingafunctionalityinthesubstrate,whichcancoordinateto
the metal complex and deliver the silicon and hydrogen substituents
to the alkyne in an intramolecular fashion [Scheme 2, eqn. (1)].9c
In the absence of such a directing group, the selective formation
of vinylsilane -2 has proven to be much more difficult; indeed
Trost and Ball only recently reported the first example of a non-
directed alkyne hydrosilylation route to this regioisomer using a
cationic Ru complex, [Cp*Ru(MeCN)3]PF6 [Scheme 2, eqn. (2)].9b
Efficient catalysts that effect the selective formation of vinylsilane,
-2 (as well as the two other stereoisomers, -(E)-1 and -(Z)-1)
therefore remain highly desirable.9a,b
Scheme 3 Application of Grubbs’catalyst 3 in hydrosilylation of ketones
and silylation of alcohols.
Prompted by these results we postulated that Grubbs’ catalyst 3
might also catalyse an alkyne hydrosilylation. We were therefore
delighted to observe that heating equimolar quantities of Et3SiH
and 1-hexyne at 60 °C, in the presence of 2.5 mol% 3 and absence
of solvent, resulted in the rapid consumption (<30 min) of both
1
starting materials (as judged by H-NMR) and the formation of
a 11:64:25 mixture of all three isomeric vinylsilane products,
-(E)-1, -(Z)-1 and -2 (entry 1, Table 1). Reducing the reaction
temperature to 22 °C (RT) led to an increased reaction time and
decrease in selectivity for the -(Z)-stereoisomer (entry 2, Table 1).
Since we observed a further reduction on continued stirring (c.f.
entries 2 and 3, Table 1), we propose that this erosion in selectivity
is most likely a result of product-scrambling on prolonged exposure
to the Ru catalyst.13
In an attempt to improve the selectivity, we carried out the
reaction in a range of solvents at 40 °C and RT. The results of this
preliminary solvent screen are summarised in Table 1. With the
exception of MeCN, which generated a range of undetermined
products, all the solvents tested proved suitable for allowing hy-
drosilylation, providing in all cases the -(Z) stereoisomer as the
major product. The fact that we could use acetone as a solvent
shows that hydrosilylation of the alkyne is much more rapid than the
Results and discussion
Over the last decade, the metathetic behaviour of Grubbs’so-called
‘first-generation’ Ru-catalyst 3 (Scheme 3) has been exploited in
† Electronic supplementary information (ESI) available: characterisation
data and NMR spectra for novel vinylsilanes. See http://www.rsc.org/
suppdata/ob/b4/b409832c/
2 5 5 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 5 5 8 – 2 5 6 2
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Article Online
Table 1 Investigating the effect of solvent, reaction concentration and temperature on the hydrosilylation of 1-hexyne with Et3SiH
Entry
Solvent
Reaction temperature/°C
Reaction concentration/M
Timea/h
% Conversionb
Ratioc -(Z)::-(E)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
no solvent
no solvent
no solvent
THF
60
—
—
—
1
1
1
1
1
1
1
0.1
0.01
10
1
0.1
1
0.1
0.01
10
1
0.5
12d
48d
2
12
2
12
2
12
2
4
8
1
4
33
1
2
8
1
100
100
100
100
100
100
100
100
100
99
64:25:11
53:12:35
37:33:30
84:16:trace
88:12:trace
67:33:trace
RT
RT
40
RT
40
RT
40
RT
40
40
40
RT
RT
RT
40
THF
CH2Cl2
CH2Cl2
MeCN
hexane
acetone
acetone
acetone
acetone
acetone
acetone
PhMe
68:18:14
e
56:11:33
89:10:1
93:6.5:0.5
100:trace:trace
80:11:9
89:9:2
95:4.5:0.5
83:15:2
96:4:trace
93:7:trace
74:22:4
78:20:2
90:10:trace
100
23
100
95
78
100
100
49
100
100
92
PhMe
PhMe
PhMe
PhMe
40
40
RT
RT
RT
2
8
PhMe
0.1
a In some instances the reaction was probably complete well before this time. b Based on complete consumption of 1-hexyne. c Ratio calculated by analysis of the
crude reaction mixture by 1H-NMR and/or by GC after the time stated in the Table. d Both starting materials were still present after 2 h but had been completely
consumed after 12 h. e Range of undetermined products in addition to the -(Z)-stereoisomer.
Table 2 Hydrosilylation of terminal alkynes using a range of silanesa
potentially competing reaction with the ketone functionality present
in the solvent (Scheme 3). Carrying out the reaction at elevated
Entry
R1
Silane
Ratiob
Yieldc (%)
temperature, instead of at RT, once again led to an increase in the
reaction rate although had relatively little effect on the selectivity.
We next focused on two rather different solvents, namely toluene
and acetone, that had performed well in our initial solvent screen,
and investigated the hydrosilylation of 1-hexyne with Et3SiH at
different concentrations and reaction temperatures. The results
are summarised in Table 1. Using both solvents, we were pleased
to observe a significant improvement in both the regio- and
stereoselectivity of the reaction, in favour of the -(Z)-product,
on decreasing the reaction concentration. Reactions performed in
toluene were more rapid than those carried out in acetone, although
our best selectivities were obtained using acetone and a 0.01 M
reaction concentration (entry 12, Table 1). However in this case, the
rate of reaction proved to be unacceptably low. When we performed
the same investigation with phenylacetylene, the reaction was
already highly selective for the -(Z)-product at 1 M in both acetone
and toluene; reducing the reaction concentration further simply
decreased the rate of hydrosilylation. Although we had previously
observed scrambling of the product stereochemistry when the
reaction had been performed in the absence of solvent (entries 1–3,
Table 1), this proved not to be a problem when the reaction was
carried under these more dilute conditions: for example, no change
in the ratio of products obtained from the reaction between 1-decyne
and PhMe2SiH (entry 10, Table 2) was observed when the reaction
mixture was stirred for a further 18 h after complete consumption of
starting materials. Whilst carrying out this study we also observed
the formation of Et3SiOH as a by-product especially when acetone
was used as solvent. However, the use of 1.1 equivalents of the
silane was sufficient to ensure complete consumption of the alkyne
starting material.
1
2d
3
C8H17
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
91:9:tr
95:5:tr
86
74
Ph
tBu
tr:tr:100
70e
f
f
4
TIPS
CO2Me
CH2OBz
CH2OTBDMS
CH2CH2OBz
CH2CH2OTBDMS
C8H17
Ph
CH2OBz
CH2OTBDMS
CH2CH2OBz
CH2CH2OTBDMS
CH2OH
C8H17
Ph
f
f
5
6g
1:97:2h
16:63:21
86:14:tr
94:6:tr
69:24:7
95:tr:5
46
51
66
90
87
81
62
73
78
79
59j
7d
8i
9
Et3SiH
10
11
12g
13d
14
15
16
17
18
19g
20g
21i
22
PhMe2SiH
PhMe2SiH
PhMe2SiH
PhMe2SiH
PhMe2SiH
PhMe2SiH
PhMe2SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
tr:93:7
6:71:23
53:28:19
66:17:17
tr:100:tr
17:73:10
>90k
f
f
CH2OBz
tr:97:3
10:80:10
5:92:3
>90k
>90k
>90k
>85k
CH2OTBDMS
CH2CH2OBz
CH2CH2OTBDMS
5:92:3l
a Reaction concentration at 0.1 M and 2.5 mol% 3 unless stated otherwise.
All reactions were generally complete within 10 h. b Ratio: -(Z)::-(E)
determined on the crude reaction mixture by 1H-NMR (tr = trace quantities
only). c Isolated yield of the vinylsilane mixture after column chromato-
graphy. d Reaction performed with 2.5 mol% catalyst and 1 M reaction
concentration. e The low yield can be attributed to loss of the volatile alkyne
f
starting material. Reaction afforded a range of undetermined products in
addition to small amounts of the desired vinylsilanes. g Reaction performed
with 5 mol% catalyst at 1 M reaction concentration. h Ratio determined on
i
reaction mixture after passing through a silica plug. Reaction performed
with 5 mol% catalyst at 0.1 M reaction concentration. j 11% of the silyl ether
vinylsilane product was also isolated. k Crude reaction yield (see Supple-
mentary Information for crude NMR spectra). l Partially overlapping vinylic
We next investigated a range of alkynes and silanes using our
optimised reaction conditions of toluene as solvent, a reaction
temperature of 40 °C and concentration of 0.1 M. The results are
summarised in Table 2. Under these conditions and with 2.5 mol%
catalyst loading, 1-decyne and phenylacetylene both reacted
smoothly with Et3SiH to provide the -(Z)-stereoisomer as the major
product (entries 1 & 2, Table 2). Reaction with 3,3-dimethylbut-1-
yne showed that bulkier alkyne substituents are also tolerated
(entry 3, Table 2), although in this case the major stereoisomer was
the -(E)-product. This presumably reflects the high energy of the
TS leading to the -(Z)-stereoisomer arising from the increased
steric hindrance provided by the bulky tBu substituent (although see
1
hydrogen resonances in the crude H-NMR for -(Z) and -(E) rendered
calculation of the relative ratio of these two isomers difficult; however the
: ratio for this entry is accurately 92:8.
mechanistic discussion below for an alternative interpretation). This
change in selectivity from the (Z)- to (E)-stereoisomer in the case
of 3,3-dimethylbut-1-yne has been observed previously.9a Mindful
of the increased length of a C–Si bond relative to a C–C bond,14
we hoped that silyl acetylenes might also undergo hydrosilylation;
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 5 5 8 – 2 5 6 2
2 5 5 9

View Article Online
however, in contrast to 3,3-dimethylbut-1-yne, TIPS-acetylene
proved to be a far poorer substrate, reacting very slowly, even in the
presence of 10 mol% catalyst and at 10 M reaction concentration,
to provide a range of undetermined products along with recovered
starting material. This suggests that electronic factors are also
important in governing the reaction efficiency. This was further
confirmed when running the alkyne into conjugation with an
electron-withdrawing ester substituent also failed to provide the
desired hydrosilylation products (entry 5, Table 2). A weaker
electron-withdrawing hydroxymethyl substituent, however, was
tolerated: the TBDMS ether of propargyl alcohol reacted well
under our standard conditions (entry 7, Table 2); the more electron-
withdrawing benzoate derivative required an increase in catalyst
loading (5 mol%) to ensure complete conversion of starting material
in a reasonable time (entry 6, Table 2). Significantly, we observed a
reversal in the regioselectivity in these two cases, with the -stereo-
isomer now predominating. By inserting an additional methylene
group between the electron-withdrawing oxygen functionality
and reacting alkyne, we were able to restore reactivity and also
selectivity for the -(Z)-stereoisomer (entries 8 & 9, Table 2).
In the case of propargyl benzoate, we propose that the proximal
carbonyl functionality acts as a directing group in the hydro-
silylation reaction.15 However, since bulky silyl protecting groups
are widely used to suppress the Lewis basicity of alcohols, the
fact that the TBDMS ether of propargyl alcohol also provides the
-stereoisomer as the major product, whereas the TBDMS ether of
homopropargylic alcohol provides the -(Z)-stereoisomer, suggests
that an electronic polarisation of the alkyne might also be capable of
governing the regioselectivity of our reaction.
Changing the silane from Et3SiH to PhMe2SiH provided similar
results although in general with reduced levels of selectivity
(entries 10–15, Table 2). Interestingly, reaction of propargyl alcohol
with PhMe2SiH provided the -stereoisomer as the major product
in 59% yield [c.f. eqn. (1), Scheme 2] along with 11% of the same
vinylsilane regioisomer in which the alcohol had also been silylated
(entry 16, Table 2).
Reactionwith(EtO)3SiHwouldprovideasyntheticallyveryuseful
vinylsilane product that can be used in Pd-catalysed cross-coupling
reactions.2 We were therefore pleased to see that this silane reacted
similarly well, although in this case, we observed a dramatic reversal
in the regioselectivity of the reaction; in all cases the -regioisomer
was obtained as the major product with good to excellent levels of
selectivity. These results therefore provide a rare example of a hy-
drosilylation reaction that is regioselective for the -stereoisomer,
irrespective of whether a directing group is present or not.
reaction that rationalises the observed selectivity differences
involving various substrates and silanes. Asking whether or not
a ruthenium alkylidene complex is the active catalyst is a useful
place to open the discussion. When we used 31P-NMR spectro-
scopy to monitor the hydrosilylation reaction between 1-hexyne
and Et3SiH in the presence of 5 mol% Grubbs’ catalyst 3, the only
major resonance observed throughout the experiment, was that for
the PCy3 ligands in the ruthenium alkylidene complex. This might
suggest that 3 provides the resting state in the catalytic cycle and
a coordinatively unsaturated ruthenium alkylidene complex 4,
generated by loss of a phosphine ligand, for example, is the active
catalyst. If this is the case, reaction might therefore commence
with the formation of the ruthenium silyl intermediate 5 through
-metathesis between the silane and the ruthenium alkylidene
4, in analogy with the classical Chauvin mechanism involving
olefins (Fig. 1).16 Subsequent insertion of the alkyne into the Ru–Si
bond would be expected to proceed in a syn fashion to provide a
vinyl ruthenium complex 6.8c,17 In the case of terminal alkynes
containing small R groups—and in analogy to related systems—
this intermediate would be expected to isomerise rapidly to the
net trans metallametalation product 7 to relieve steric interactions
between the silyl group and ruthenium complex.8c,17b,c,e -Hydride
elimination would then provide the -(Z)-stereoisomer vinylsilane
product and regenerate the ruthenium alkylidene complex 4 thereby
completing the catalytic cycle. If the rate of isomerisation of 6 to
7 is low, as would be expected for alkynes containing sterically
demanding substituents such as 3,3-dimethyl-but-1-yne, then direct
-hydride elimination from the syn metallametalation intermediate
6 would nicely account for the formation of the -(E) stereoisomer
for this class of alkyne substrates. Further evidence for the formation
of a vinyl ruthenium intermediate such as 7 comes from our
identification of a silylacetylene by-product 8, which we observed
in trace amounts (<1%) by GC-MS analysis of the products derived
from the reaction between Et3SiH and 1-decyne. The formation of
this by-product can be rationalised by invoking a competing -
hydride elimination on the vinyl ruthenium isomerisation product 7
forming the silylacetylene 8,17a and a hydrido ruthenium complex 9,
which itself may also be catalytically active.
Obviously our 31P-NMR experiment does not discount the
possibilitythatverysmallconcentrationsofanotherclassofruthenium
complex, which were not detectable by NMR, are mediating the
reaction. Thus an alternative possibility is that decomposition of
Grubbs’ catalyst 3 generates a coordinatively unsaturated species
10 that then undergoes a more standard oxidative addition with the
silane to generate an intermediate H–Ru–Si species 11 (Fig. 2).17a
The thermal decomposition pathways of ruthenium alkylidene
complexes have been investigated by Grubbs and proposed to
It is interesting to speculate on the nature of the active
catalyst in our hydrosilylation reaction and a mechanism of
Fig. 1 A possible mechanism of hydrosilylation involving Grubbs’ catalyst.
2 5 6 0
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 5 5 8 – 2 5 6 2

View Article Online
Fig. 2 Alternative mechanistic pathways could also account for the stereoselectivity of the hydrosilylation reaction.
involve initial phosphine dissociation (also required for metathesis)
although the inorganic decomposition products have not been
elucidated.18 Furthermore the non-metathetic behaviour of Grubbs’
catalyst 3, for example in olefin isomerisation,19 is often attributed
to the formation of ruthenium hydride species through decomposi-
tion of the alkylidene pre-catalyst.20 With a H–Ru–Si species 11 in
hand, two possibilities for further reaction now exist. According to
the classical Chalk–Harrod mechanism for hydrosilylation,21 alkyne
insertion into the Ru–H bond generates the vinyl ruthenium inter-
mediate 12 through a net syn hydrometalation process in which the
bulkier ruthenium centre adds to the less hindered alkyne terminus
(in analogy to most hydrometalation reactions involving terminal
alkynes). Subsequent reductive elimination would then afford the -
(E)-vinylsilane stereoisomer.17a However, since this isomer is only
ever formed in trace quantities, with the exception of 3,3-dimethyl-
but-1-yne, the alternative pathway involving initial alkyne insertion
into the Ru–Si bond should be considered as an alternative reaction
pathway. In this case, and as before, regioselective syn addition
of the Si–Ru bond across the alkyne provides a vinyl ruthenium
complex 13.17a Subsequent isomerisation to 14 as described above,
followed by reductive elimination, would account for the formation
of the -(Z)-vinylsilane stereoisomer. This mechanism would also
explain the formation of the silylacetylene by-product 8 in the same
way as that outlined in Fig. 1.
substituent. In these cases, we suggest that direct anti metalla-
metalation (as postulated by Trost)22 of the longer (than Ru–H)
Ru–Si bond across an alkyne coordinated to the Ru centre through
the proximal oxygen donor group, would generate a relatively stable
five-membered ruthenaoxacycle 15, which on reductive elimination
wouldprovidethe-regioisomer(Fig.3).Incorporatinganadditional
methylene group between the oxygen donor and reacting alkyne
functionality returns the selectivity in favour of the -regioisomer
(Table 2, entries 8 and 9) on account of i) the reduced stability of
the analogous six-ring ruthenaoxacycle that would be formed in the
6-endo cyclisation mode;23 ii) the availability of a more favourable
5-exo cyclisation mode; iii) location of the bulkier silyl substituent
at the less hindered terminus of the alkyne (Fig. 3).
The putative mechanisms described in Figs. 1 and 2 are
stepwise in the way in which the silane and alkyne substrates
react at the metal centre. A rather different mechanistic pathway
for cationic ruthenium complexes has recently been proposed:22
density functional calculations carried out by Trost suggested that
oxidative addition of the Si–H bond to the ruthenium centre proceeds
in a concerted fashion along with hydride insertion to the metal-
activated alkyne. The calculations nicely accounted for the
experimentally observed anti addition of the silane across the alkyne,
as well as the observed Markovnikov addition regiochemistry.9b It
may be the case that such a pathway is also be operating in our
system although it is not clear how this mechanism would account
for the differing regioselectivity obtained between Et3SiH and
(EtO)3SiH. We believe this may be accounted for in another way.
We have already suggested that terminal alkynes derived from
propargyl alcohol provide the -regioisomeric alkyne preferentially
on account of the directing nature of the proximal oxygen
Fig. 3 Proximal donor groups affect the regioselectivity of hydro-
silylation.
In the case of (EtO)3SiH, which favours the -regioisomer
irrespective of whether or not a directing group is present in the
alkyne substrate (Table 2, entries 17–22), we tentatively suggest
that an oxygen atom in one of the ethoxy ligands on the silane
provides a similar coordinating effect (Fig. 4). In this case, alkyne
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 5 5 8 – 2 5 6 2
2 5 6 1

View Article Online
1375–1408; (c) J. S. Panek, in Comprehensive Organic Synthesis, eds.
B. M. Trost and I. Fleming, Pergamon, Oxford, 1991, vol. 1, pp 579–
627; (d) I. Fleming, J. Dunoguès and R. Smithers, Org. React. (N.Y.),
1989, 37, 57–575; (e) T. A. Blumenkopf and L. E. Overman, Chem.
Rev., 1986, 86, 857–873; (f ) T. H. Chan and I. Fleming, Synthesis, 1979,
761–786; (g) Y. Ie, N. Chatani, T. Ogo, D. R. Marshall, T. Fukuyama,
F. Kakiuchi and S. Murai, J. Org. Chem., 2000, 65, 1475–1488.
2 (a) S. E. Denmark, R. F. Sweis and D. Wehrli, J. Am. Chem. Soc., 2004,
126, 4865–4875; (b) S. E. Denmark and R. F. Sweis, J. Am. Chem. Soc.,
2004, 126, 4876–4882; (c) S. E. Denmark and M. H. Ober, Aldrichimica
Acta, 2003, 36, 75–85; (d) S. E. Denmark and R. F. Sweis, Acc. Chem.
Res., 2002, 35, 835–846; (e) T. Hiyama and E. Shirakawa, Top. Curr.
Chem., 2002, 219, 61–85.
3 (a) S. E. Denmark and S.-M. Yang, Org. Lett., 2001, 3, 1749–1752;
(b) A. G. M. Barrett, J. C. Beall, D. C. Braddock, K. Flack, V. C. Gibson
and M. M. Salter, J. Org. Chem., 2000, 65, 6508–6514; (c) S. Chang and
R. H. Grubbs, Tetrahedron Lett., 1997, 38, 4757–4760.
4 (a) C. Pietraszuk, B. Marciniec and H. Fischer, Tetrahedron Lett.,
2003, 44, 7121–7124; (b) C. Pietraszuk, H. Fischer, M. Kujawa and
B. Marciniec, Tetrahedron Lett., 2001, 42, 1175–1178; (c) A. K.
Chatterjee, J. P. Morgan, M. Scholl and R. H. Grubbs, J. Am. Chem.
Soc., 2000, 122, 3783–3784.
5 (a) K. Oshima, in Science of Synthesis, ed. I. Fleming, Georg Thieme
Verlag, NewYork, 2002, vol. 4, pp 713–756; (b) F. Kakiuchi,A. Yamada,
N. Chatani, S. Murai, N. Furukawa and Y. Seki, Organometallics, 1990,
18, 2033–2036; (c) F. Kakiuchi, Y. Tanaka, N. Chatani and S. Murai, J.
Organomet. Chem., 1993, 456, 45–47.
6 (a) T. Hiyama and T. Kusumoto, in Comprehensive Organic
Synthesis, eds. B. M. Trost and I. Fleming, Pergamon, Oxford, 1991,
vol. 8, pp 769–773; (b) J. L. Speier, Adv. Organomet. Chem., 1979, 17,
407–447.
Fig. 4 Coordinating groups on the silane may also govern the regio-
selectivity of hydrosilylation.
insertion into the Ru–Si bond gives rise to two five-membered
ruthenaoxasilacycle products 16 and 17 depending on the regio-
selectivity of insertion; since that (16) leading to the -regioisomer
suffers from unfavourable steric interactions between the ligands at
the ruthenium centre and the alkyne substituent, a more favourable
pathway via 17 is followed thereby accounting for the preferential
formation of the -regioisomer with this silane.
In summary, we have shown that commercially available
Grubbs’ first-generation metathesis catalyst can be used to mediate
the hydrosilylation of terminal alkynes. In no case did we observe
the presence of cross-metathesis products on analysis of the crude
reaction mixture by GC-MS. The reaction selectivity displays
an interesting reaction concentration dependence and more
significantly, the choice of silane governs the regioselectivity.
Although we have speculated upon possible mechanisms for this
hydrosilylation, future work now needs to be directed towards
validating these mechanistic hypotheses. Deuterium labelling
studies, a thorough kinetics analysis of the reaction, and detailed
NMR experiments, in addition to the identification of side-
products by careful GC-MS analysis of the reaction mixture, will
help shed light on our postulated mechanisms and allow us to
identify favoured pathways. The commercial availability of Grubbs’
catalyst 3, and a number of related ruthenium alkylidene complexes,
combined with their ease of handling, bodes well for the widespread
uptake of these reagents for effecting hydrosilylation reactions. To
this end, we now need to screen other metal alkylidene catalysts for
hydrosilylation activity and further investigate the synthetic scope
of the reaction by investigating other alkynes (including internal
alkynes) and silanes.
7 J. L. Speier, J. A. Webster and G. H. Barnes, J. Am. Chem. Soc., 1957,
79, 974–979.
8 For some recent examples involving Rh complexes: (a) A. Mori,
E. Takahisa, Y. Yamamura, T. Kato, A. P. Mudalige, H. Kajiro,
K. Hirabayashi, Y. Nishihara and T. Hiyama, Organometallics, 2004,
23, 1755–1765; (b) R. Takeuchi and N. Tanouchi, J. Chem. Soc., Perkin
Trans. 1, 1994, 2909–2913; (c) I. Ojima, N. Clos, R. J. Donovan and
P. Ingallina, Organometallics, 1990, 9, 3127–3133; (d) For a compre-
hensive review: I. Ojima, Z. Li and J. Zhu, in The Chemistry of Organo-
silicon Compounds, eds. Z. Rappoport and Y. Apeloig, John Wiley and
Sons, Chichester, 1998, Volume 2, Chapter 29, pp 1687–1792.
9 For some recent examples involving Ru complexes: (a) Y. Kawanami,
Y. Sonoda, T. Mori and K. Yamamoto, Org. Lett., 2002, 4, 2825–2827;
(b) B. M. Trost and Z. T. Ball, J. Am. Chem. Soc., 2001, 123, 12726–
12727; (c) Y. Na and S. Chang, Org. Lett., 2000, 2, 1887–1889.
10 (a) T. M. Trnka and R. H. Grubbs, Acc. Chem. Res., 2001, 34,
18–29; (b) A. Fürstner, Angew. Chem., Int. Ed., 2000, 39, 3012–3043;
(c) R. H. Grubbs and S. Chang, Tetrahedron, 1998, 54, 4413–4450;
(d) S. K. Armstrong, J. Chem. Soc., Perkin Trans. 1, 1998, 371–388.
11 B. Alcaide and P. Almendros, Chem. Eur. J., 2003, 9, 1258–1262.
12 S. V. Maifeld, R. L. Miller and D. Lee, Tetrahedron Lett., 2002, 43,
6363–6366.
Experimental
General procedure for hydrosilylation
Grubbs’ catalyst 3 (2.5 or 5 mol%) was added in one portion to
a solution of the alkyne (1.0 equiv.) and the silane (1.2 equiv.)
in toluene (1 M or 0.1 M) and the reaction mixture was heated
at 40 °C. TLC monitoring indicated that the reactions were
generally complete within 10 h. Removal of the solvent under
reduced pressure and purification by silica gel chromatography
(hexane–Et2O eluent) afforded the desired mixture of vinylsilane
products as a colourless oil. No attempts were made to separate the
isomeric mixtures.
13 In subsequent reactions stopping the reaction as soon as all starting
material had been consumed minimised this problem.
14 C–Si bond length ~1.85 Å; C–C bond length ~1.54 Å.
15 (a) T. Murai, F. Kimura, K. Tsutsui, K. Hasegawa and S. Kato, Organo-
metallics, 1998, 17, 926–932; (b) G. Stork, M. E. Jung, E. Colvin and
Y. Noel, J. Am. Chem. Soc., 1974, 96, 3684–3686.
16 (a) J. L. Herrisson and Y. Chauvin, Makromol. Chem., 1970, 141, 161–
176; (b) for a recent discussion of the mechanism of olefin metathesis
see: C. Adlhart and P. Chen, J. Am. Chem. Soc., 2004, 126, 3496–3510.
17 (a) H. Katayama, K. Taniguchi, M. Kobayashi, T. Sagawa, T. Minami
and F. Ozawa, J. Organomet. Chem., 2002, 645, 192–200; (b) S. M.
Maddock, C. E. F. Rickard, W. R. Roper and L. J. Wright, Organo-
metallics, 1996, 15, 1793–1803; (c) C.-H. Jun and R. H. Crabtree, J. Or-
ganomet. Chem., 1993, 447, 177–187; (d) M. A. Esteruelas, J. Herrero
and L. A. Oro, Organometallics, 1993, 12, 2377–2379; (e) R. S. Tanke
and R. H. Crabtree, J. Am. Chem. Soc., 1990, 112, 7984–7989.
18 M. Ulman and R. H. Grubbs, J. Org. Chem., 1999, 64, 7202–7207.
19 B. Schmidt, Synlett, 2004, 1541–1544 and references therein.
20 B. Schmidt, Eur. J. Org. Chem., 2004, 1865–1880.
Acknowledgements
We thank The University of Birmingham for a studentship
(to C. S. A.), Pfizer (unrestricted award to L. R. C.) and a referee
for some very useful suggestions regarding possible mechanisms
of hydrosilylation.
21 A. J. Chalk and J. F. Harrod, J. Am. Chem. Soc., 1965, 87, 16–21.
22 L. W. Chung, Y.-D. Wu, B. M. Trost and Z. T. Ball, J. Am. Chem. Soc.,
2003, 125, 11578–11582.
References
1 (a) I. Fleming, A. Barbero and D. Walter, Chem. Rev., 1997, 97,
2063–2192; (b) E. Langkopf and D. Schinzer, Chem. Rev., 1995, 95,
23 J. M. Brown, Chem. Soc. Rev., 1993, 22, 25–41.
2 5 6 2
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 5 5 8 – 2 5 6 2