Homogeneous catalysis. Use of a ruthenium(ii) complex for catalysing the ene
reaction
William W. Ellis, W. Odenkirk and B. Bosnich*†
Department of Chemistry, The University of Chicago, 5735 S. Ellis Avenue, Chicago, IL 60637, USA
+
The complex trans-[Ru(salen)(NO)(H
2
O)] catalyses the ene
after reaction, the homoallylic alcohol product is expected to be
labile for efficient turnover.
Some of these results are collected in Table 1. As is
commonly observed for Lewis acid promotion, only electron-
difficient carbonyl-containing enophiles engage in the ene
reaction at acceptable rates. Generally, ruthenium(ii) aquo
reaction between activated enophiles and olefins to give
homoallylic alcohols by a stepwise process.
The oxo-ene reaction, which usually involves the addition of
alkenes to aldehydes to produce homoallylic alcohols [eqn. (1)],
complexes are weak acids having a pK
(
a
similar to AcOH acid
in water).13 In order to determine the ability of acid to promote
H
H
(1)
the reactions, 2 mol% TFA was used under the same conditions
to promote the reaction shown in entry 1 (Table 1). It was found
that < 1% conversion to the ene product occurred after 150 h,
indicating that acid is neither responsible for, nor competitive
with, the ruthenium catalysed process.
Although, like other promoters, the ruthenium catalyst has its
limitations for the ene reaction, it serves to demonstrate that
structurally defined transition metal complexes which are not
normally oxophilic can be modified to act as genuine catalysts
for the generally resistant ene reaction. It is probable, however,
that the present catalyst will find practical applications in the
intramolecular ene reaction. The conversion of (+)-citronellal to
l-isopulegol [eqn. (2)] is an important step in the industrial
O
H
O
R
R
is potentially a useful method of carbon–carbon bond construc-
tion. Its implementation, however, suffers from a number of
difficiencies. Among these is the necessity of employing high
temperatures for most reactions. The obvious method for
promoting the reaction at lower temperatures by using conven-
1
2
3
4
5
3 3 4 4
tional Lewis acids such as AlCl , BF , SnCl or TiCl , can
1
lead to new problems. Whereas these Lewis acids do indeed
promote the reaction, they lead to the formation of Lewis acid
alcohol intermediates which can release a proton. The released
acid can itself act as a catalyst (Prins reaction) and can cause
other reactions to occur with the substrates and product.
Snider1 has shown that the use of Me
,6
2
AlCl can promote the
(
2)
ene reaction without proton interference because the methyl
groups are capable of rapidly scavenging the proton to give
methane. The resultant formation of alkoxide adducts, however,
consumes the Lewis acid promoter and, consequently, the
Lewis acid is used in stoichiometric or greater quantities. In
addition, alkylaluminium halides have been shown to transfer
H
O
OH
production of l-menthol14 where ZnBr
2
in stoichiometric
amounts at ca. 5 °C is used to give l-isopulegol in 95% yield
over the other isomers. Using 1 mol% of the ruthenium catalyst
in MeNO at 25 °C, (+)-citronellal is converted to l-isopulegol
2
after 6 h. An 80% yield of l-isopulegol was obtained, with the
remaining product consisting of the other (three) isomers.
Given this efficient conversion, it is possible to entertain the
prospect of using chiral analogues of the present catalyst for
asymmetric catalytic intramolecular ene reactions.
In order to ascertain if the present catalysed ene reactions are
concerted or stepwise processes we have investigated the
products from a number of 1,3-dienes. A typical transformation
is illustrated in eqn. (3), together with the putative inter-
7
the alkyl group to the aldehyde and to undergo the Oppenauer
8
II9
oxidation. Milder Lewis acids such as those derived from Zn
and Ti
IV10
have been shown to act as true catalysts for a limited
number of intramolecular and intermolecular ene reactions,
respectively.
9
,14,15
Here we report the use of a new type of Lewis acid catalyst
6
for the ene reaction. It is the d ruthenium(ii) complex, trans-
[
Ru(salen)(NO)(H
2
O)] SbF
6
1, which we have employed for
+
OH2
N
O
N
Ru
+
C6F5
Ru
N
O
O –
OH
O
H
1
C6F5
catalysing the Diels–Alder11 and Mukaiyama reactions.12 The
+
Ru+]
[
60%
air and moisture stable catalyst 1 is readily prepared in two steps
(
2 mol%)
0 °C
CD3NO2
] and the ligand.11
from commercially available [Ru(NO)Cl
3
+
(3)
5
II
H
O
+
Although Ru complexes are generally electron-rich and
consequently do not act as Lewis acids, the incorporation of the
C6F5
Ru
–
5
h
O
O
+
electron-withdrawing NO ligand, the coordination of hard
+
ligands, namely N and O, and the presence of a positive charge
all conspire to produce a Lewis acidic ruthenium centre.
F5
C6F5
40%
+
Further, because the water ligand trans to the NO ligand is very
labile,11 this coordination site is readily available for binding
the enophile (usually an aldehyde) to activate it for reaction and,
mediates. The two products are formed in constant kinetic
proportions and are not interconverted in the presence of the
Chem. Commun., 1998
1311
Ellis, William W.
