ric amounts have been used in 1,2-reduction to prevent alkene
hydroboration.7 DIP-chloride was reported in 1,2-reduction
of 2-cyclohexen-1-one, but after 7 days the alcohol was
isolated in only 36% ee.8 Biotransformations using enzymes
are perhaps the most common methods used in asymmetric
1,2-reduction of enones. Unfortunately, this method typically
requires long reaction times and large excess of the enzyme
performing the reduction.9,10 Additionally, the R,ꢀ-unsatur-
ated ketone must be soluble in water and usually only one
enantiomeric alcohol can be synthesized by enzyme route.
Chiral deprotonation of epoxides has been explored as an
alternative route toward chiral allylic alcohols, but this
method is also highly substrate limited.11 Our interest in the
asymmetric reduction of prochiral ketones using the chiral
boronic ester TarB-NO2 (Figure 1) and NaBH4 prompted us
to explore the asymmetric reduction of R,ꢀ-unsaturated
ketones as a potential route to chiral allylic alcohols.
method did favor 1,2-reduction in TarB-NO2 mediated
reaction with 2-cyclohexen-1-one, the amount of 2 was still
significant and very little asymmetric induction was observed
(entry 4). Previous research in our laboratory had demon-
strated the ability of lithium aminoborohydride (LAB)
reagents to selectively reduce the carbonyl of both R,ꢀ-
unsaturated aldehydes and ketones.14 However, in TarB-NO2-
mediated reduction lithium pyrrolidinoborohydride yielded
a 75:25 mixture of 1 and 2 and nearly racemic alcohol (entry
5).
Table 1. TarB-NO2-Mediated Reduction of 2-Cyclohexen-1-onea
reaction
temp (°C)/time
entry
hydride
NaBH4
NaBH(OAc)3
NaBH(OPh)3
NaBH4/B(OH)3
LiBH3(pyrrolidine)
1:2b
% ee 1c
1
2
3
4
5
25/30min
25:75
100:0
80:20
87:13
75:25
33
33
17
6
0/12
0/12
h
h
Figure 1. L-TarB-NO2.
25/30min
25/3
h
6
a General reaction conditions: 1 mmol of ketone dissolved in 2 mL of
0.5 M TarB-NO2 (1 mmol) followed by 2 mmol of hydride. b Ratio
determined by GC. c Enantiomeric excess determined by chiral GC
Our initial studies utilized NaBH(OAc)3 as a stoichiometric
reducing agent as this reagent had been reported to give
predominantly allylic alcohol products.12 We expected that
substituting NaBH4 with NaBH(OAc)3 in TarB-NO2 medi-
ated reduction would allow us to enantioselectively reduce
R,ꢀ-unsaturated ketones to the corresponding chiral allylic
alcohols. We also included other reducing agents, such as
NaBH4, NaBH(OPh)3, and LiBH3(pyrrolidine). Accordingly,
TarB-NO2 was mixed in equimolar amounts with 2-cyclo-
hexen-1-one and 2 equiv of the hydride source (Table 1).
Initial reduction of 2-cyclohexen-1-one with TarB-NO2 and
NaBH4 at room temperature gave a mixture of 2-cyclohexen-
1-ol (1) and 2-cyclohexan-1-one (2) in a 25:75 ratio (entry
1) and the desired allylic alcohol was obtained in 33% ee.
Reduction of the same ketone using NaBH(OAc)3 at 0 °C
for 12 h showed complete regioselectivity in the reduction,
but the asymmetric induction was similar to that obtained
with NaBH4 (entry 2). Substitution of NaBH(OAc)3 with
NaBH(OPh)3 led to a decrease in asymmetric reduction and
a 80:20 mixture of 1 and 2 (entry 3). A recent publication
described the use of NaBH4 and boric acid in the selective
1,2-reduction of R,ꢀ-unsaturated ketones.13 Although this
The low enantioselectivity in TarB-NO2-mediated reduc-
tion of 2-cyclohexen-1-one led us to consider substrate
modification to enhance asymmetric induction. Our compu-
tational modeling predicted that the transition state was
lowest in energy when the carbonyl carbon was proximal to
the carboxylic acid moiety of TarB-NO2.15 However, our
model was unclear in delineating the influence of electronics
and sterics in the asymmetric induction involving TarB-NO2
mediated asymmetric reduction. Our recent work on asym-
metric reduction of aliphatic ketones suggested that steric
requirements of the alkyl groups attached to the carbonyl
functionality were key in achieving high induction.16 For
example, the TarB-NO2 mediated asymmetric reduction of
2-octanone gave the product alcohol in 60% ee whereas
pinacolone, a ketone containing two sterically distinct alkyl
groups, gave the product in 95% ee. We also noted that
2-methyl-3-pentanone gave product alcohol in 62% ee,
similar to the results obtained with 2-octanone. Apparently,
TarB-NO2 reagent does not significantly distinguish an
isopropyl group from an ethyl group.
(7) Cho, B. T. Tetrahedron 2006, 62, 7621.
(8) Brown, H. C.; Ramachandran, P. V. Acc. Chem. Res. 1992, 25, 16.
In a recent report on the synthesis of allocolchicine
analogues, TarB-NO2-LiBH4 was used in the reduction of
ketone 3.1 However, the product alcohol 4 was obtained in
(9) Pollard, D. J.; Telari, K.; Lane, J.; Humphrey, G.; McWilliams, C.;
Nidositko, S.; Salmon, P.; Moore, J. Biotechnol. Bioeng. 2006, 93, 674
.
(10) Zagozda, M.; Plenkiewicz, J. Tetrahedron: Asymmetry 2006, 17,
1958
.
(11) (a) Oxenford, S. J.; Wright, J. M.; O’Brien, P.; Panday, N.; Shipton,
M. R. Tetrahedron Lett. 2005, 46, 8315. (b) Gayet, A.; Andersson, P. G
AdV. Synth. Catal. 2005, 347, 1242.
(14) Pasumansky, L.; Goralski, C. T.; Singaram, B. Org. Process Res.
DeV. 2006, 10, 959.
(12) Gribble, G. W. Chem. Soc. ReV. 1998, 27, 395.
(13) Cho, B. T.; Kang, S. K.; Kim, M. S.; Ryu, S. R.; An, D. K.
Tetrahedron 2006, 62, 8164.
(15) Cordes, D. B.; Nguyen, T. M.; Kwong, T. J.; Suri, J. T.; Luibrand,
R. T.; Singaram, B. Eur. J. Org. Chem. 2005, 5289.
(16) Kim, J.; Singaram, B. Tetrahedron Lett. 2006, 47, 3901.
Org. Lett., Vol. 11, No. 19, 2009
4359