ChemComm
Cite this: Chem. Commun., 2011, 47, 3989–3991
COMMUNICATION
Sequential Birch reaction and asymmetric Ir-catalyzed hydrogenation as
a route to chiral building blocksw
Alexander Paptchikhine,a Kaori Ittob and Pher G. Andersson*ac
Received 16th December 2010, Accepted 26th January 2011
DOI: 10.1039/c0cc05619g
A range of 1,2,4-trisubstituted cyclohexadienes obtained from
the Birch reaction were hydrogenated asymmetrically to produce
synthetically valuable chiral compounds in high enantio- and
diastereoselectivity.
asymmetric hydrogenation by our library of N,P-chelated Ir
catalysts.6
The best catalysts were [A-IrCOD]BArF and [B-IrCOD]-
BArF (Fig. 1, BArF
= tetrakis(3,5-bis(trifluoromethyl)-
phenyl)borate). In contrast to our previous observations on
benzyl enol ethers,7 the methoxy enol ethers studied here were
hydrogenated without noticeable hydrolysis.
The Birch reaction is an old and well-studied transformation
that converts aromatic compounds into non-conjugated
dienes.1 It has been used to synthesize precursors for many
natural products and other important building blocks.2 The
rich and varied chemistry of aromatic compounds, combined
with the powerful functionalisation that is provided by the
Birch reduction, makes the latter an important tool for
producing complex molecules.
Yields from the Birch reactions and the results of the
subsequent asymmetric hydrogenations are summarized in
Table 1. For most substrates, high trans-to-cis ratios
(470 : 30) were observed, and the trans isomers were hydro-
genated in excellent ee values (94–99%). Low ee values were
observed for the minor cis isomers.
Most interestingly, substrate 5 was hydrogenated with
exceptional regio- (499 : 1) and stereoselectivity (499.9%,
Scheme 1). The tetrasubstituted double bond remained intact,
leaving the obtained chiral product open for further modifi-
cations. For example, oxidation with RuCl3/NaIO4 produced
compound 7 (Scheme 1).
Ir-catalyzed asymmetric hydrogenation has recently
emerged as an efficient route towards enantiomerically
enriched hydrocarbons.3 To date, asymmetric hydrogenation
has not been applied to products of the Birch reaction,
although these compounds are frequently subjected to other
transformations that produce molecules with new chiral
centers. There are also many reports of chiral centers being
introduced during the Birch reaction by means of chiral
auxiliaries,4 or simply by using chiral aromatic substrates.5
The Birch reaction’s high regioselectivity and tolerance for
substitution make it an excellent source of prochiral, unsatu-
rated substrates for asymmetric hydrogenation.
To add another layer of selectivity, the catalysts could
be deactivated towards methoxy-substituted olefins using
poly(4-vinylpyridine) (PVP)8 as an additive. This allowed
compounds 4e (69% yieldz) and 4m–o to be prepared
i
(Scheme 2). Other bases such as pyridine, NEt3, Pr2NEt,
aniline, Ph3P, 1,10-phenanthroline, and Cy2NH (5 eq. with
respect to the catalyst) completely poisoned the catalyst and
discolored reaction solution. Results using KOAc as a base
were comparable to those with PVP. Compound 4o could be
easily hydrolyzed by an acid, then equilibrated under basic
conditions9 to give a disubstituted cyclohexanone which can
be used in the preparation of a 7-membered lactone, in high ee
(Scheme 2).
A range of cyclohexadienes were prepared from aromatic
compounds using Na or Li in liquid ammonia (Table 1). In the
case of compounds 1a and 1f, a near-stoichiometric amount of
tert-butanol was used as the proton source, and a co-solvent
was also used. The more hindered substrates 1c–1e and 1g–o
required a large excess of Na and ethanol for acceptable
conversion. The obtained dienes were purified by distillation
or column chromatography, then screened as substrates for
The trans preference observed for substrates 2a, 2c–l and 5
suggests that, after reduction of the first double bond, the
substrate dissociates from the catalyst before undergoing the
second reduction. The catalyst selects which olefin face to bind
and hydrogenate based on sterics, as was previously shown in
our selectivity model.6b
a Department of Biochemistry and Organic Chemistry,
Uppsala University, Box 576, S-75123 Uppsala, Sweden.
E-mail: pher.andersson@biorg.uu.se; Fax: +46 18-471-3818;
Tel: +46 18-471-3816
b Graduate School of Life Sciences, Tohoku University,
980-8577 Sendai, Japan
As shown in Scheme 3, the ee of the preferred trans isomer is
enriched at the expense of the ee of the minor cis isomer.
Whenever the catalyst reduces the double bond on the ‘wrong’
face, producing 8a (which could be reduced to the minor trans
enantiomer), the mistake is quickly corrected by the fast
reduction of 8a to the cis isomer. The result is a poor ee for
c School of Chemistry, University of KwaZulu-Natal, Durban,
South Africa
w Electronic supplementary information (ESI) available: Experimental
details, separation methods and spectral data. See DOI: 10.1039/
c0cc05619g
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 3989–3991 3989