Formation of allylsilanes from Cl2[P(C6H11)3]2Ru=C(R)H and vinylsilanes - β-SIR'3 elimination from ruthenacyclobutanes as a terminating step in olefin cross-metathesis

Article abstract of DOI:10.1039/b008536g

Stoichiometric reactions of the Grubbs carbene complex [Cl₂{P(C₆H₁₁)₃}₂Ru=C(Ph)H] with vinylsilanes, H₂C=C(Si-Me(n)R(3-n))H (R = Ph, OEt; n = 1, 2, 3), afford metathesis products and allylsilanes formed by β-SiR₃ elimination followed by reductive elimination; the formation of allylsilanes constitutes a terminating step in the Ru-catalysed cross-metathesis of olefins with methylsubstituted vinylsilanes.

Full text of DOI:10.1039/b008536g

Formation of allylsilanes from Cl2[P(C6H11)3]2RuNC(R)H and
vinylsilanes—b-SiRA3 elimination from ruthenacyclobutanes as a terminating
step in olefin cross-metathesis
Cezary Pietraszuka,b and Helmut Fischer*a
a Fachbereich Chemie, Universität Konstanz, Fach M727, D-78457 Konstanz, Germany.
E-mail: hfischer@dg6.chemie.uni-konstanz.de
b Faculty of Chemistry, Adam Mickiewicz University, Grunwaldzka 6, 60-780 Poznan, Poland
Received (in Cambridge, UK) 23rd October 2000, Accepted 8th November 2000
First published as an Advance Article on the web
Stoichiometric reactions of the Grubbs carbene complex
[Cl2{P(C6H11)3}2RuNC(Ph)H] with vinylsilanes, H2CNC(Si-
MenR32n)H (R = Ph, OEt; n = 1, 2, 3), afford metathesis
products and allylsilanes formed by b-SiR3 elimination
followed by reductive elimination; the formation of allyl-
silanes constitutes a terminating step in the Ru-catalysed
cross-metathesis of olefins with methylsubstituted vinyl-
silanes.
Only compounds 3a and 9a are those expected from olefin
metathesis. All other products (4a–8) are allyl derivatives
containing the C3 fragment of the metallacycle. Obviously, the
metallacycle formed by addition of 2 to the RuNC bond of 1a
breaks down by two different pathways: (a) reductive decou-
pling to give the metathesis products 3a, 9a and a [Ru]NC(R)H
species and (b) elimination of an allylsilane derivative and
formation of a [Ru] fragment.
Pathway (b) causes a decrease of the catalytically active
[Ru]NC(R)H species and presumably is the most important
factor in reducing the efficiency of the catalytic system. The
conclusion is supported by the following observation: When Me
in 2 is replaced stepwise by OEt, both the ratio of the metathesis
product (type 3) to the allyl derivative (type 4) in the
stoichiometric reaction [0.26 (SiMe3), 0.68 (Si(OEt)Me2), 13
(Si(OEt)2Me)] and the cross-metathetical conversion of styrene
to vinylsilane mixtures catalysed by 1 drastically increase.
Analogously, a strong shift toward the metathesis product is
observed when Me is stepwise displaced by Ph.
Metallacyclobutanes1 play an important role in a number of
stoichiometric and catalytic transformations. Two of the
catalytically most important reaction modes of metallacyclo-
butanes are (a) reductive elimination to give cyclopropanes and
a metal–ligand fragment2 and (b) reductive decoupling to form
an olefin–carbene complex (olefin metathesis).3 In these
reactions the metallacyclobutanes are formed as intermediates
by addition of the CNC bond of an olefin to the MNC bond of an
LnMNC(R)RA complex. Several types of olefin metathesis are
known, such as ring-opening metathesis polymerisation, self-
and cross-metathesis of linear olefins, acyclic diene metathesis
(ADMET), and ring-closing metathesis (RCM).3
The reactions of 2 with the methylidene complex 1b and the
ethylidene complex 1c instead of 1a proceeded similarly, albeit
more slowly. The reaction rate decreased in the series 1a > 1b
We recently reported4 on the highly selective cross meta-
thesis of styrene H2CNC(Ph)H, with several vinylsilanes
H2CNC(SiR3)H, to give (E)-H(Ph)CNC(SiR3)H and ethylene
catalysed by the Grubbs catalyst [Cl2{P(C6H11)3}2RuNC(Ph)H]
(1a).5 Very high conversions even at rt were observed when R
= ORA (RA = Et, SiMe3); however, the conversion significantly
decreased with increasing substitution of Me for ORA.4 To
determine the reason for the decreasing selectivity we studied
the stoichiometric reaction of 1a with various vinylsilanes and
now report on (a) the first evidence for b-silyl migration in
metallacyclobutanes, (b) the very high selectivity of b-SiR3
versus b-H migration and (c) hints for b-migration as the
terminating step thus limiting the efficiency of the catalyst.
When an equivalent of trimethylvinylsilane (2) was added to
a solution of 1a in C6D6 a smooth reaction was observed.† After
6 h at rt 98% of 2 and 99% of 1a had been consumed. A detailed
analysis of the organic reaction products revealed the formation
of 15% of 3a, 57% of 4a, 6% of 5a, in addition to small amounts
of 6 (5%), 7 (2%), 8 (5%), 9a (2–5%) (Scheme 1) and
unidentified Ru complexes. Cyclopropanes were not detected.
>
1c. Again mixtures of metathesis products and allyl
derivatives were obtained. The reactions of 2 with 1b and 1c
were accompanied by a substrate-independent decomposition
of the Ru complexes6 which gave rise to a reduction in the
conversion of 2 (41% for 1b and 35% for 1c, each after 6 h).
Within error limits, the product distribution after 6 h was the
same as that after 18 h.
The formation of the major allyl derivative (4a) in the
reaction of 1a with trimethylvinylsilane (2) can be explained in
two different ways: (i) by b-SiMe3 elimination to give an
allyl(silyl) complex (see Scheme 2: C/D) followed by reductive
elimination or (ii) by b-H elimination to give a hydrido(a-
silylallyl) complex (Scheme 2: E/F) again followed by
reductive elimination. The labeling experiment (Scheme 2)
allows exclusion of pathway (ii). The 2H-NMR spectrum of the
products obtained from the reaction of 1a with H2CNC(D)SiMe3
(2-d1) exhibited only signals in the olefinic region. From the
absence of signals in the aliphatic region it follows that 4aA-d1
has not been produced. The formation of more than 1% of 4aA-d1
(with respect to 4aA-d1) would have been detected. Another
product of pathway (ii), compound 10, has not been identified
among the reaction products in earlier experiments (see Scheme
1). These results indicate that there is a strong preference for b-
SiR3 elimination over b-H elimination. In fact, products derived
from b-H elimination have not been detected in the stoichio-
metric experiments. Thus the ratio 4a/5a = 57+6 presumably
reflects the relative stabilities of different allyl(silyl) complex
intermediates C and D.
The bis(silyl)allyl derivatives 6 and 7 are secondary products
derived from the reaction of [Ru]NC(SiMe3)H with 2. Their
formation establishes that [Ru]NC(SiMe3)H species are also
formed in the reaction of 1 with 2 as we proposed earlier.4 Until
Scheme 1
DOI: 10.1039/b008536g
Chem. Commun., 2000, 2463–2464
This journal is © The Royal Society of Chemistry 2000
2463

efficiency of the system. The migratory aptitude considerably
decreases in the series SiMe3 > Si(OEt)Me2 > Si(OEt)2Me and
SiMe3 > Si(Ph)Me2 > Si(Ph)2Me. Although there is a strong
preference for SiR3 migration compared to H migration it seems
likely that b-H elimination and subsequent reductive elimina-
tion in systems without a SiR3 substituent could also drain the
active species from the catalytic cycle and thus limit the turn-
over number in cross-metathesis.
C. P. acknowledges a research grant from the Alexander von
Humboldt Foundation. We also thank Professor Ulrich Groth
for supplying access to GC-MS and Dr A. Geyer for 2H-NMR
spectra. Financial support of this work by the Fonds der
Chemischen Industrie is gratefully acknowledged.
Notes and references
† Typically, in an NMR tube 1.21 3 1025 mol of vinylsilane was added by
syringe to a solution of [Cl2{P(C6H11)3}2RuNC(Ph)H] (0.01 g, 1.21 3 1025
mol) and anthracene (internal standard) dissolved in 0.65 ml of C6D6. The
reactions were followed by 1H-NMR spectroscopy for 6 h. Conversions and
selectivities were calculated using the internal standard method.7 Products
1
were identified by GC-MS spectra and by a comparison of their H NMR
spectra and their retention times (GC) with those of authentic samples.
1 For a representative review on group 8 metallacyclobutanes see W. P.
Jennings and L. L. Johnson, Chem. Rev., 1994, 94, 2241.
2 See e.g. M. P. Doyle, Chem. Rev., 1986, 86, 919; M. Brookhart and W. B.
Studabacker, Chem. Rev., 1987, 87, 411; H.-W. Fruehauf, Chem. Rev.,
1997, 97, 523.
3 For recent reviews see: K. J. Ivin and J. C. Mol, Olefin Metathesis and
Metathesis Polymerization, Academic Press, San Diego, 1997; Alkene
Metathesis in Organic Synthesis, ed. A. Fuerstner, Springer, Berlin, 1998;
R. H. Grubbs and S. Chang, Tetrahedron, 1998, 54, 4413.
4 C. Pietraszuk, B. Marciniec and H. Fischer, Organometallics, 2000, 19,
913.
5 P. Schwab, M. B. France, J. W. Ziller and R. H. Grubbs, Angew. Chem.,
1995, 107, 2179; Angew. Chem., Int. Ed. Engl., 1995, 34, 2039; P.
Schwab, R. H. Grubbs and J. W. Ziller, J. Am. Chem. Soc., 1996, 118,
100.
Scheme 2
now, all our attempts to synthesise [Cl2{P(C6H11)3}2RuNC(Si-
Me3)H] or to spectroscopically detect RuNC(SiMe3)H species in
the cross-metathesis of vinylsilanes with styrene catalysed by 1
have failed.4
Our results demonstrate that b-SiR3 elimination in b-SiR3-
substituted ruthenacyclobutanes followed by reductive elimina-
tion strongly competes with olefin metathesis and thus pre-
sumably is the most important factor limiting the catalytic
6 M. Ulman and R. H. Grubbs, J. Org. Chem., 1999, 64, 7202.
7 Quantitative Analysis using Chromatographic Techniques, ed. E. Katz,
Wiley, Chichester, 1987.
2464
Chem. Commun., 2000, 2463–2464