Chiral base mediated transformation of cyclic 1,3-diketones

Article abstract of DOI:10.1039/b606864b

Treatment of certain 1,3-diketones with a chiral lithium amide base results in the formation of a non-racemic lithium mono-enolate; these intermediates can be transformed directly into chiral hydroxyketone products by reduction with DIBAL-H in high yield and with selectivities of up to 99% ee. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b606864b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Chiral base mediated transformation of cyclic 1,3-diketones
Benjamin Butler, Thomas Schultz and Nigel S. Simpkins*
Received (in Cambridge, UK) 15th May 2006, Accepted 4th July 2006
First published as an Advance Article on the web 27th July 2006
DOI: 10.1039/b606864b
Although these preliminary findings validated our initial plan
Treatment of certain 1,3-diketones with a chiral lithium amide
base results in the formation of a non-racemic lithium mono-
enolate; these intermediates can be transformed directly into
chiral hydroxyketone products by reduction with DIBAL-H in
high yield and with selectivities of up to 99% ee.
for accessing chiral cyclic hydroxyketones, we experienced
significant problems in generalising the chemistry, due to the
highly sensitive nature of the enol silane intermediates. Substantial
improvement to the overall efficiency of the method could be
achieved by the simple expedient of using the crude enol silanes
(usually contaminated by 5–10% of starting diketone) in the
reduction. Adopting this procedure enabled the enantioselective
formation of the hydroxyketones 8–13 in overall yields of 69–79%
from starting dione, and with promising levels of asymmetric
induction.
The enantioselective reduction of prochiral cyclic diketones to give
chiral, non-racemic hydroxyketone derivatives is usually accom-
plished by Baker’s yeast,1,2 although other microorganisms have
also been used,3,4 and recently chiral oxazaborolidine catalysts
have also been employed.5
Based on our recent results, which have demonstrated the utility
of chiral lithium amide base enolisation for asymmetric transfor-
mation of cyclic imides,6 we anticipated that a similar type of
enolisation would be possible for cyclic diketones. We also
expected that this type of enolisation would facilitate overall
enantioselective reduction of a diketone, either directly or via enol
silane intermediates. These ideas have proved to have some
foundation, and we show herein that highly selective chiral base
enolisation of cyclic diketones is indeed possible, and that this
provides a new method of asymmetric diketone reduction.
In preliminary explorations, selective enolisation of a cyclopen-
tanedione 1, by addition of the chiral base 2, using Me3SiCl as the
electrophilic trapping agent, gave mono-enol silane 3 in moderate
yield (50%) and enantiomeric excess (65% ee), Scheme 1.
In these reductions high levels of diastereoselectivity were
observed, with reduction occurring to give mainly the isomer
shown (.95 : 5 dr), which results from hydride addition syn to the
smaller C-2 substituent (i.e. the methyl group in each case). This is
significant in that the aforementioned bio-transformation methods
often give problematic mixtures of isomers.
This initial enolisation–Me3SiCl trapping process establishes a
stereogenic centre at C-2 and sets the scene for subsequent
diastereoselective reduction of the remaining carbonyl function.
Thus, reduction of the ketone function of enol silane 3 using
DIBAL-H, with concomitant enol silane hydrolysis on work-up,
gave hydroxyketone 4, with similar levels of enantiomeric
enrichment to 3. We then achieved similar results for the analogous
cyclohexanedione 5; correlation of products 4 and 7 with
compounds described by Brooks and co-workers enabled the
assignment of relative and absolute stereochemistry.1
Although we were encouraged by the new results described
above, two steps (pots) are required to accomplish a single selective
reduction process. The formation of an intermediate enol silane
acts as a device for selectively protecting one of the ketone
functions present in the starting dione. We proposed that a one-
pot variant might be possible if, instead of preparing intermediate
enol silanes, we simply utilised the initially formed chiral lithium
mono-enolate 14 as the reduction substrate, Scheme 2.
The use of a derived metal enolate to protect a ketone function
in a molecule, whilst another carbonyl is reduced, is an established
(although not widely used) strategy for regio- or chemoselective
reduction,7 but the idea has not been explored before in
conjunction with a chiral base process. When applied to our
diketone substrates we obtained the results shown in Table 1.{
Scheme 1
School of Chemistry, The University of Nottingham, University Park,
Nottingham, UK NG7 2RD.
E-mail: Nigel.Simpkins@Nottingham.ac.uk; Fax: +44(0) 115 951 3564;
Tel: +44(0) 115 951 3533
Scheme 2
3634 | Chem. Commun., 2006, 3634–3636
This journal is ß The Royal Society of Chemistry 2006

View Article Online
Table 1 Synthesis of 8–13 according to Scheme 2
Entry Product Yield (%) Product de (%)a Product ee (%)b
1
2
3
4
5
6
a
8
9
10
11
12
13
69
69
71
77
81
70
31
99
58
95
35
98
77
99
74
99
67
88
Fig. 1 Sense of enolisation using base 2.
has not been explored in detail and at this stage it is not possible to
make generalisations concerning the scope, yields and selectivities
of the process.
Determined by 1H NMR spectroscopy and cross-checked with data
b
.
b
from
column except for 8—by GC using a 2,3-di-O-pentyl-c-CD column.
Determined by HPLC using either a Chiralcel OD or OJ
It is worth noting that the observed sense of enantioselectivity
seen in the enolisation of these diketones is that expected based on
precedent for simple cyclic ketones—nicely illustrated by compar-
ing the site of proton abstraction in dione 15 with that in
4-alkylcyclohexanones, Fig. 1.
Diastereomer ratios were also evident from these analyses.
Two key differences were observed in the results obtained this
way, compared to the reductions involving the enol silane
intermediates. Firstly, for the five-membered systems (odd-
numbered entries) the moderate levels of asymmetric induction
are maintained, whilst the almost complete diastereoselectivity seen
before is seriously eroded. More significantly, for the six-
membered systems (even-numbered entries) the diastereoselectivity
of the reduction is maintained, and the overall enantioselectivity is
significantly enhanced—hydroxyketones 9 and 11 being formed in
essentially enantiomerically pure form.
Therefore it is possible to use the new chemistry in a predictive
sense to synthesise useful chiral ketone building blocks.
In conclusion, we have described a new variant of the chiral base
enantioselective enolisation, applicable to cyclic diones, which
enables either one-pot or two-pot overall asymmetric reduction. As
shown in Scheme 3, there is also potential for a powerful
desymmetrisation involving C–C bond formation, and we expect
that this aspect can be developed by correct choice of nucleophilic
organometallic.
The erosion of diastereoselectivity in the reduction of the five-
membered lithium enolates, but not for the six-membered cases, is
not straightforward to explain. In the reduction of lithium enolates
14, issues of enolate aggregation, and/or the formation of (chiral)
intermediate aluminium ‘ate’-complexes, could be responsible for
the observed effects.
We are grateful to the Engineering and Physical Sciences
Research Council (EPSRC) for a studentship (to BB) and a
Fellowship (to TS).
Notes and references
In terms of enantiomeric excess, the intrinsic enantioselectivity
of the chiral base in the deprotonation step appears to be more
effectively translated into the ee of the hydroxyketone using the
‘enolate protecting group’ method. When using the enol silanes,
high selectivity is probably undermined by the extreme sensitivity
of these intermediates, which results in varying degrees of
hydrolysis (on handling or in situ) that leads ultimately to racemic
hydroxyketone.
{ Typical procedure e.g. preparation of (2)-11: the chiral amine
hydrochloride salt (73 mg, 0.28 mmol) was suspended in dry THF (2 ml),
cooled to 278 uC, and a solution of n-butyllithium in hexane (2.11 M,
0.26 ml) was added. The mixture was stirred at room temperature for
10 min. 2-Benzyl-2-methylcyclohexane-1,3-dione 15 (50 mg, 0.23 mmol)
was dissolved in dry THF (2 ml) and cooled to 278 uC. The base was
cooled to 278 uC and added to the diketone via transfer cannula. The
mixture was stirred at 278 uC for 1 h, before a solution of DIBAL-H in
THF (1.7 M, 0.45 ml) was added. The mixture was stirred at 278 uC for
4 h, before the reaction was quenched by the addition of HCl (1 M, 2 ml).
The reaction mixture was diluted with ethyl acetate (20 ml) and water
(3 ml), and the organic phase was separated. The aqueous phase was
extracted with ethyl acetate (3 6 5 ml) and the combined organic phases
were washed with HCl (1 M, 3 6 3 ml), brine (3 6 3 ml) and dried over
MgSO4. The solvent was removed under reduced pressure, and the residue
was purified by column chromatography (silica, EtOAc–petroleum ether,
1 : 3, Rf = 0.23) giving the desired ketoalcohol 11 as a white solid, mp 78–
Our success in achieving hydride addition to the non-racemic
enol silanes and the mono-lithium enolate intermediates 14
prompted us to attempt an analogous Grignard addition.
Initially we tested this idea using allylmagnesium bromide, e.g.
Scheme 3.
Thus, generation of the chiral enol silane 16, starting from dione
15, followed by Grignard addition, gave the hydroxyketone 17 in
good yield and ee, and as a single diastereoisomer. The alternative,
direct, method, involving addition of the Grignard reagent to the
chiral base reaction mixture, gave the same level of induction, 17
being isolated in slightly lower yield (74%). So far this chemistry
1
79 uC, (39 mg, 0.18 mmol, 77%), [a]D 225 (c 0.5 in CH2Cl2). H NMR
(400 MHz, CDCl3): 1.08 (3H, s, CH3), 1.72–1.91 (2H, m, CH2), 1.98–2.19
(2H, m, CH2), 2.51–2.59 (2H, m, CH2), 2.97 (1H, d, J 14, CHHPh), 3.11
(1H, d, J 14, CHHPh), 3.77 (1H, dd, J 7, 3, CHOH), 7.14–7.33 (5H, m,
Ar). 13C NMR (100 MHz, CDCl3) 20.4 (CH3), 20.7 (CH2), 28.5 (CH2),
37.3 (CH2), 37.7 (CH2), 54.5 (C), 75.7 (CH), 126.3 (CH), 128.0 (CH), 130.6
(CH), 137.5 (C), 213.8 (C). IR (solution in CH2Cl2): 3687, 3603, 2942, 2685,
2410, 2302, 1705, 1604, 1516, 1494, 1373, 1065, 992, 969 cm21. MS (EI,
180 uC) m/z (%): 218 (62), 159 (14), 147 (19), 127 (16), 117 (12), 99 (11), 91
(100), 71 (8). HR-MS (EI, 180 uC): Calc. for C14H18O2: 218.1307. Found:
218.1307. HPLC: (Chiralcel OD, hexane–IPA, 95 : 5, 0.5 ml min21):
41.6 min (major), 54.9 min (minor); 27.0 min and 38.1 min (minor
diastereomer); 99% ee, 95% de.
1 D. W. Brooks, H. Mazdiyasni and P. G. Grothaus, J. Org. Chem., 1987,
52, 3223.
2 Z.-L. Wei, Z.-Y. Li and G.-Q. Lin, Tetrahedron: Asymmetry, 2001, 12,
229.
3 Z.-L. Wei, Z.-Y. Li and G.-Q. Lin, Synthesis, 2000, 1673.
Scheme 3
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 3634–3636 | 3635

View Article Online
4 K. Shimoda, N. Kubota, H. Hamada and H. Hamada, Tetrahedron
Lett., 2006, 47, 1541.
D. J. Bennett, P. L. Pickering and N. S. Simpkins, Chem. Commun.,
2004, 1392.
5 M. Shimizu, S. Yamada, Y. Fujita and F. Kobayashi, Tetrahedron:
Asymmetry, 2000, 11, 3883.
6 (a) D. J. Adams, A. J. Blake, P. A. Cooke, C. D. Gill and N. S.
Simpkins, Tetrahedron, 2002, 58, 4603 and references therein; (b)
7 (a) D. H. R. Barton, R. H. Hesse, M. M. Pechet and C. Wiltshire, J.
Chem. Soc., Chem. Commun., 1972, 1017; (b) D. H. R. Barton, R. H.
Hesse, M. M. Pechet and C. Wiltshire, J. Chem. Soc., Perkin Trans. 1,
1977, 1075; (c) G. A. Kraus and K. Frazier, J. Org. Chem., 1980, 45, 4262.
3636 | Chem. Commun., 2006, 3634–3636
This journal is ß The Royal Society of Chemistry 2006