The coupling of methylene and vinyl ligands at a ruthenium(ii) centre
Anthony F. Hill,* Chean T. Ho and James D. E. T. Wilton-Ely
Department of Chemistry, Imperial College of Science Technology and Medicine, South Kensington, London, UK SW7 2AY
The reaction of [Ru(CHNCH2)Cl(CO)(PPh3)2] with diazo-
Maitlis’ studies of C–C bond forming processes for organorho-
dium species certainly point to a greater migratory aptitude for
vinyl than phenyl ligands.1 However, since the phenyl complex
2a has been shown to be in equilibrium with [Ru(Ph)Cl(NCH2)-
(CO)(PPh3)2]6 the relative mobility of vinyl and phenyl ligands
appears not to be the limiting factor here, but more likely one of
product development control.
3
methane
provides
the
allyl
complex
[Ru(h -
CH2CHCH2)Cl(CO)(PPh3)2] which is also the product of the
reactions of [RuHCl(CO)(PPh3)3] with either allene or
propyne and which reacts subsequently with NaS2CNMe2 to
provide [RuH(S2CNMe2)(CO)(PPh3)2] and allene.
In a recent review,1 Maitlis et al. drew attention to the potential
significance of the coupling of vinyl and methylene groups in
the Fischer–Tropsch synthesis. In particular the oxidation of
[Rh2(m-CH2)2(CHNCH2)2(h-C5Me5)2] by AgBF4 in acetonitrile
Complex 3 also results from two other approaches. Santos
and coworkers have shown that in general, terminal alkynes
RC·CH (R = Ph, But, Bun) react with [RuHCl(CO)(PPh3)3] to
provide coordinatively unsaturated s-vinyl complexes
[Ru(CHNCHR)Cl(CO)(PPh3)2].10 We anticipated that a similar
hydroruthenation of allene might provide 3 which indeed it does
in high yield (Scheme 1). Somewhat surprisingly however, 3 is
also the product of the reaction of [RuHCl(CO)(PPh3)3] with
propyne! This result can be explained by the formation of the
usually less favoured a-substituted s-vinyl complex [Ru(C-
MeNCH2)Cl(CO)(PPh3)2], which can undergo b-Ru–H elimina-
3
was shown to provide [Rh(h -CH2CHCH2)(NCMe)(h-
C5Me5)]BF4,2 whilst diazomethane addition to the diruthenium
complex [Ru2(m-CHNCH2)(CO)3(h-C5H5)2]+ followed by de-
protonation provides [Ru2(m-CHCHNCH2)(m-CO)(CO)(h-
C5H5)2].3 The operation of these couplings on bimetallic
systems, whilst clearly significant in understanding Fischer–
Tropsch processes, begs the question do they occur on
monometallic systems? Werner and coworkers have reported
that the reaction of [RuCl(NCPh2)(PPh3)(h-C5H5)] with
2
tion to provide [RuH(h -CH2CCH2)Cl(CO)(PPh3)2] followed
by hydroruthenation to provide 3.
3
vinyl magnesium bromide provides the allyl complex
The chemistry of h -allyl ligands is enriched by their ability
3
[Ru(h -CH2CHCPh2)(PPh3)(h-C5H5)].4
The
complex
to assume a monohapto coordination mode. We therefore
attempted to induce monohapto coordination by replacing the
chloride ligand in 3 with a potentially bidentate dithiocarbamate
ligand. Surprisingly, the reaction of 3 with Na[S2CNMe2] in
dichloromethane–ethanol led to the formation of [RuH(S2CN-
Me2)(CO)(PPh3)2] 4 with the organometallic ligand being lost
from the coordination sphere. When the reaction was carried out
in CDCl3–MeOH (10:1), 1H NMR integration confirmed allene
as the organic product. The most reasonable course for this
reaction would appear to be dissociation of chloride followed by
the b-RuH elimination of allene to provide the coordinatively
unsaturated, though presumably solvent-stabilised species
‘[RuH(CO)(PPh3)2]+’. Although this type of elimination is
[Ru(CHNCH2)(NCPh2)(PPh3)(h-C5H5)] is an attractive inter-
mediate, however as the authors point out, the product is also
consistent with direct nucleophilic attack at the carbene carbon.
Such a process is typical of coordinatively saturated carbene
complexes of divalent ruthenium.5 Here, we report the unambi-
guous observation of the coupling of a vinyl and a methylene
ligand on a ruthenium system. Furthermore, we show an
3
unusual b-elimination process of the resulting h -allyl ligand
which may be induced by dithiocarbamate coordination.
Roper and coworkers have shown that the coordinatively
unsaturated complex [Ru(Ph)Cl(CO)(PPh3)2] 1a reacts with
diazomethane to provide a remarkably stable methylene
2
complex [Ru(h -OCPh)Cl(NCH2)(PPh3)2] 2a, but that such a
unusual,
we
have
observed11
that
treating
reaction has so far failed for the corresponding trans b-styryl
analogue [Ru(CHNCHPh)Cl(CO)(PPh3)2] 1b.6 An intriguing
feature of the reactivity of 2a is the apparent reluctance of the
phenyl and methylene ligands to couple, despite the anticipated
stability of the hypothetical product [Ru(CH2Ph)Cl-
(CO)(PPh3)2] 1c. We find that the parent vinyl complex
[Ru(CHNCH2)Cl(CO)(PPh3)2] 1d7 reacts readily with diazo-
methane in dichloromethane–diethyl ether to provide high
yields of the h -allyl complex [Ru(h -CH2CHCH2)Cl(CO)-
(PPh3)2] 3 (Scheme 1), the identity and stereochemistry of
which follows unambiguously from spectroscopic data† and
two unequivocal syntheses (vide infra). Whilst the reaction
PPh3
H
C
PPh3
CH2
ii
CH2
OC
Cl
OC
Cl
Ru
Ru
C
H
CH2
PPh3
2b
PPh3
1d
ii
i
3
3
PPh3
Cl
OC
Ph3P
Ph3P
iii or iv
H
Ru
Ru
Ph3P
Cl
PPh3
C
O
leading to 3 is unique, the class of compounds has precedent.
3
Related
examples,
e.g.
[Ru(h -CH2CHCH2)Cl(CO)-
3
iv
(PMe2Ph)2] have been obtained from the reaction of [RuCl2-
iii
(CO)2(PMe2Ph)2] with Bun SnCH2CHNCH2,8‡ whilst [RuHCl-
3
(CO)(PPh3)3] hydroruthenates 1,3-dienes to provide substituted
examples.9 Although this result is surprising, the methylene
complex [Ru(CHNCH2)Cl(NCH2)(CO)(PPh3)2] 2b is an ob-
vious intermediate in the formation of 3. It remains unclear to us
why, in the case of 1a methylene coordination is followed by
(reversible) migration of a phenyl group to a carbonyl ligand,
but in the case of 1d, migration of the vinyl ligand occurs
selectively and irreversibly to the introduced methylene ligand.
PPh3
PPh3
Ru
Me
H
OC
Cl
OC
C
CH2
Ru
C
H2C
Cl
CH2
PPh3
PPh3
Scheme 1 Reagents and conditions: i, HC·CH (CH2Cl2, 25 °C); ii, CH2N2
(Et2O, CH2Cl2, 25 °C); iii, CH3C·CH (thf, 25 °C then heating); iv,
H2CNCNCH2 (thf, 25 °C then heating)
Chem. Commun., 1997
2207
Hill, Anthony F.
