Enantioselective reduction of ketones with NaBH4/diglyme possibly catalysed by trialkyl borate: Optically active sec-alcohols from prochiral ketones with catalytic (-)-menthol: Autocatalysis option

Article abstract of DOI:10.1016/j.tetasy.2004.11.078

Ketones can be reduced with NaBH₄ in diglyme without an apparent proton source, but putatively catalysed by a trialkyl borate. This can be initially derived in situ from NaBH₄ and an alcohol, although the reaction becomes increasingly autocatalytic with time. With (-)-menthol as the initiating alcohol, the enantioselective reductions of a range of prochiral ketones in quantitative yields and moderate ees (generally 58-87%) were realised (the autocatalysis was demonstrated in two cases; catalysis by tris-(-)-menthyl borate was demonstrated in one case). The mechanism may involve either the activation of the substrate (electrophilically) or of the hydride reagent (nucleophilically). This method offers a relatively simple and inexpensive approach to a key transformation in asymmetric synthesis.

Full text of DOI:10.1016/j.tetasy.2004.11.078

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 751–754
Enantioselective reduction of ketones with NaBH4/diglyme possibly
catalysed by trialkyl borate: optically active sec-alcohols from
prochiral ketones with catalytic (À)-menthol: autocatalysis option
Sosale Chandrasekhar* and Raghunandan Hota
Department of Organic Chemistry, Indian Institute of Science, Bangalore 560 012, India
Received 25 October 2004; accepted 15 November 2004
Available online 21 January 2005
Abstract—Ketones can be reduced with NaBH4 in diglyme without an apparent proton source, but putatively catalysed by a trialkyl
borate. This can be initially derived in situ from NaBH4 and an alcohol, although the reaction becomes increasingly autocatalytic
with time. With (À)-menthol as the initiating alcohol, the enantioselective reductions of a range of prochiral ketones in quantitative
yields and moderate ees (generally 58–87%) were realised (the autocatalysis was demonstrated in two cases; catalysis by tris-(À)-
menthyl borate was demonstrated in one case). The mechanism may involve either the activation of the substrate (electrophilically)
or of the hydride reagent (nucleophilically). This method offers a relatively simple and inexpensive approach to a key transformation
in asymmetric synthesis.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
NaBH4 reduction of ketones may apparently be pro-
moted either by the electrophilic activation of the sub-
strate or the nucleophilic activation of the reagent—or,
of course, both. This, in fact, leads to a possible asym-
metric variant of the borohydride reaction, catalysis by
an optically active alcohol in an inert solvent. The
design of asymmetric variants of the complex metal
hydride reductions has attracted much interest and con-
tinues to be a challenge.7–12
The reduction of ketones with sodium borohydride
(NaBH4) is an important transformation in organic syn-
thesis, which is characterised by its effectiveness and sim-
plicity. It is generally believed to require a proton
source, usually water or an alcohol such as ethanol, that
is employed as a solvent for the reaction.1–4 The role of
the proton source is mechanistically unclear: impor-
tantly, the activation of the carbonyl group via proton-
ation at the ketone oxygen atom may be ruled out, given
the large difference in the pKaÕs of ketones (ꢀÀ7) and
alcohols (ꢀ+16).5 Also, NaBH4 reacts competitively
with the hydroxylic solvent employed (to produce
borate species)—and yet, the reduction reaction works
excellently in practice.1–4
2. Results and discussion
We explored the above possibility in diethyleneglycol di-
methyl ether (diglyme, an aprotic solvent, which has
been recommended1 for NaBH4). When a variety of pro-
chiral ketones 1 was treated with a solution of NaBH4 in
dry diglyme, which had been pre-treated with 0.05 M
equiv (relative to NaBH4) of (1R)-menthol, the corres-
ponding alcohols 2 were produced over a period of
24–40 h, in essentially quantitative yields and generally
moderate ees (Scheme 3 and Table 1; note that the ees
are P80% in two cases). Also, no detectable reaction
occurred without the above pre-treatment with the
menthol. The scope of the reaction is demonstrated by
its success with aliphatic, alicyclic and aromatic
substrates (Table 1).
An interesting possibility is that the reaction is catalysed
by the trialkyl borate I, which is formed as a by-product
(Scheme 1); in fact, the reducing species may also be the
trialkoxyborohydride IV (Scheme 2). It is known that
borohydride reactions are electrophilically catalysed by
metal ions (Li+ or Mg2+),4 and that alkoxyborohydrides
are better hydride donors than NaBH4 itself.6 Thus, the
*
Corresponding author. Tel.: +91 80 2293 2689; fax: +91 80 23600
529; e-mail: sosale@orgchem.iisc.ernet.in
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.11.078

752
S. Chandrasekhar, R. Hota / Tetrahedron: Asymmetry 16 (2005) 751–754
NaBH4 + 4 ROH
[NaBH4-n(OR)n]
B(OR)3 + RONa + 4H2
(i)
I
B(OR)3 + RONa
I
B(OR)4 Na+
II
(ii)
B(OR)3
O
BH3
+ B(OR)3
B(OR)3
+ BH3
O
O
R
R'
R'
R'
R
R
H
H
H
BH3
IV
III
(iii)
Scheme 1. Possible catalysis by trialkyl borate I of the reduction of a ketone with NaBH4. Borate I is produced by the reaction of added alcohol with
NaBH4, as in (i), and subsequently may form the Ôate complexÕ II, as in (ii): this equilibrium, however, would considerably favour I as NaBH4 is in
large excess relative to ROH (catalyst). Possible electrophilic catalysis by I of the reduction reaction is shown in (iii). The III converts to IV as BH3 is
more electrophilic than B(OR)3.7 Further reaction occurs with IV and its higher alkoxy analogues (instead of NaBH4), under catalysis by B(OR)3.
BH3 + NaB(H)(OR)3
(iv)
B(OR)3 + NaBH4
Na+
V
ONa
H
O
B(OR)3
Na+
O
R'
R
R
R'
R'
R
H
H
+ B(OR)3
B(OR)3
VI
V
(v)
(vi)
BH3 + VI
IV + B(OR)3
Scheme 2. Possible reduction of a ketone by a trialkoxyborohydride V. This may be formed by the reaction of a trialkyl borate with NaBH4, as in
(iv); V may transfer hydride to a ketone more efficiently than NaBH4 does itself, as in (v); for the reaction in (vi), cf. Scheme 1; IV may react with
B(OR)3 as in (iv) to regenerate V.
Mechanistically, however, the reaction sequences in
Schemes 1 and 2 would need to be modified in the case
of the above reactions initiated by menthol: even if it
presumed that the reaction is initially catalysed by tris-
(1R)-menthyl borate, the trialkyl borate derived from
the product alcohol, which is indeed the primary prod-
uct prior to work-up, would itself act as a catalyst.
Clearly, as the reaction proceeds, it would increasingly
be autocatalysed. This is apparently advantageous in
that the reaction can be initiated by the product alcohol
itself (presuming its availability), thus eliminating the
need to separate it from menthol or any other initiating
alcohol that may be employed.
OH
O
(i)
1/2
a
b
c
d
e
f
R
Ph
p-
R'
Me
Me
Me
R
R
R'
Tolyl
R'
H
p-Anisyl
(99%)
1
2
Naphth-2-yl Me
n-Hexyl Me
α−Tetralone
(i) NaBH4 (1.3 eq.)/(-)-menthol (0.05 eq.)/
diglyme/N2/25 oC/24-40 h
Scheme 3. Enantioselective reduction of the prochiral ketones 1 to the
corresponding secondary alcohols 2, catalysed by (À)-menthol (cf.
Table 1).
On the other hand however, an autocatalytic process
severely limits the possibilities for stereocontrol (as the
chiral auxiliary is invariable), and the observed levels
of induction would have to be accepted as they are. In
other words, a chiral autocatalytic reaction would be
useful only if the induction is Ônaturally highÕ. It should
be noted that the product alcohol may not always be
available, so that the initiation may have to be per-
formed with an available chiral alcohol, for example,
(1R)-menthol. Interestingly, it would be possible to
avoid autocatalysis, if the initiating trialkyl borate is
The assumption that the reactions are possibly catalysed
by the trimenthyl borate is supported by the observation
that the reduction of acetophenone 1a, with NaBH4 in
diglyme in the presence of tris-(1R)-menthyl borate,
afforded the corresponding alcohol 2a in 69% ee. Also,
chiral catalysis by an alcohol other than (À)-menthol
was demonstrated by the reduction of naphthyl ketone
1d in the presence of the (À)-octanol 2e: it is noteworthy
that different enantiomers of 2d were obtained in two
cases (entries 5 and 6).

S. Chandrasekhar, R. Hota / Tetrahedron: Asymmetry 16 (2005) 751–754
753
Table 1. Enantioselective reduction of the ketones 1 to the alcohols 2
with NaBH4/diglyme, catalysed by (À)-menthol (Scheme 3)a
or THF). However, the reactions are mechanistically un-
clear and their scope limited with the chiral auxiliary
also being used at stoichiometric levels. Presumably,
the reductions are mediated by an alkoxyborohydride
species, for example, HnB(RCO2)4–n, where RCO2 repre-
sents L-tartrate, cf. Scheme 2. Thus, the present method
offers considerable improvements in terms of both pro-
cedure and scope.
Entry Ketone 1 Time Product 2 Configuration– Ee (%)b
(h)
rotation
1
2
3
4
5
6
7
8
1a
1a
1b
1c
1d
1d
1e
1f
24
24
24
24
36
36
40
30
2a
2a
2b
2c
2d
2d
2e
2f
1S-(À)
1S-(À)
1S-(À)
1S-(À)
1S-(À)
1R-(+)
2R-(À)
1R-(À)
62 (62)
58
63 (63)
23 (43)
80 (74)
63
87
58 (57)
3. Conclusion
a Entries 2 and 7 refer to autocatalysis: the reaction was performed not
with (À)-menthol, but with added 2a and 2d, respectively; entry 6
refers to catalysis by (À)-octan-2-ol 2e instead of (À)-menthol.
b Polarimetrically determined values, with the corresponding chiral
HPLC values in parenthesis. Typical procedure: NaBH4 (1.3 mmol)
in dry diglyme (5 mL) was treated with (À)-menthol (0.05 mmol),
under dry N2. The (effervescent) mixture was stirred for 15 min/
25 ꢁC, and treated with prochiral ketone 1 (1.0 mmol, in one portion)
for 24–40 h; the progress of the reaction was followed by TLC. The
mixture was worked-up by quenching with saturated NH4Cl solution
(5 mL), concentrating in vacuo to remove most of the diglyme,
adding water (20 mL) and extracting the mixture with Et2O, etc. The
resulting crude product 2 was purified by flash column chromato-
graphy (SiO2/eluent: 20% EtOAc–hexane). The resulting pure 2a–f
were identified spectrally (IR, 300 MHz 1H NMR and MS); the
specific rotations were determined (on a JASCO DIP-370 digital
polarimeter), and compared with reported values to obtain the
configurations and the ees.13–15 It was demonstrated that the reaction
of 1a could be catalysed by tris-(À)-menthyl borate16 (0.2 equiv)
rather than by (À)-menthol itself: a nearly quantitative yield of 2a in
69% ee was realised.
In conclusion, we have developed an asymmetric version
of the well-known NaBH4 reduction of ketones that is
catalytic in the chiral auxiliary: when the enantiopure
alcohol product is available, the process can be per-
formed entirely autocatalytically. The yields are quanti-
tative and the ees generally moderate (good in two
cases). The process also offers insights into the mecha-
nism of the reaction of NaBH4 with ketones (notably,
a plausible explanation for the necessity of a Ôproton
sourceÕ). Further work to elaborate the interesting
results obtained is planned.
Acknowledgements
We thank the CSIR (New Delhi) for generous financial
support of this work. We are most grateful to Prof. S.
Chandrasekaran for access to the HPLC facility in his
group, without which this work could not have been
completed.
The polarimetric ee values were also confirmed by chiral HPLC
analysis, except in the case of octanol 2e, which could not be
resolved.
A cellulose carbamate based Chiralcel OD column
(250 · 4.6 mm) was employed on a Shimadzu LC-10AS instrument,
eluting with hexane–isopropanol (90:10 v/v) at a flow rate of 0.8 mL/
min, and monitoring at 260 nm (315 nm in the case of 2d).
References
1. Banfi, L.; Narisano, E.; Riva, R. In Encyclopedia of
Reagents for Organic Synthesis; Paquette, L. A., Ed.; John
Wiley: Chichester, 1995; Vol. 7, pp 4522–4528, and
references cited therein.
2. March, J. Advanced Organic Chemistry, 4th ed.; John
Wiley: New York, 1992; pp 910–918, and references cited
therein.
far more electrophilic than the product trialkyl borate.
The autocatalytic possibility has been demonstrated in
the case of two substrates, acetophenone and octan-2-
one (58% and 87% ee, respectively); it is clear that only
the octanone case is viable.
3. Greeves, N. In Comprehensive Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 8, pp
1–24, and references cited therein.
4. House, H. O. Modern Synthetic Reactions; W.A. Benja-
min: Menlo Park (California), 1972; pp 47, 51, and
references cited therein.
5. Ref. 2, pp 250–252, and references cited therein.
6. Wigfield, D. C. Tetrahedron 1979, 35, 449–462.
7. Itsuno, S. In Comprehensive Asymmetric Catalysis; Jacob-
sen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer:
Berlin–Heidelberg, 1999; Vol. 1, pp 289–315, and refer-
ences cited therein.
8. Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry
of Organic Compounds; John Wiley: New York, 1994; pp
942–947, and references cited therein.
Asymmetric reductions of ketones by complex metal hy-
dride reagents have attracted much attention by several
groups in recent years, although apparently with mixed
results.7–12 The results have been particularly disap-
pointing so far in the case of NaBH4, although an
LiBH4/N-benzoylcysteine reagent is quite efficient.9
The related oxazaborolidine-based reductions are both
efficient and act at catalytic levels,7–9 but are less acces-
sible than NaBH4. Moreover, other unsaturated func-
tionalities would interfere to varying extents with these
borane based reactions. The present work employing
the relatively simple and highly chemoselective NaBH4,
represents an important departure from previous ap-
proaches to the enantioselective reduction of ketones.
9. Nishizawa, M.; Noyori, R. In Comprehensive Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon:
Oxford, 1991; Vol. 8, pp 159–182, and references cited
therein.
10. Yatagai, M.; Ohnuki, T. J. Chem. Soc., Perkin Trans. 1
1990, 1826–1828.
Previous approaches have employed NaBH4 modified
by a chiral acid or alcohol, for example L-tartaric acid10
or a monosaccharide,11,12 in an aprotic solvent (benzene

754
S. Chandrasekhar, R. Hota / Tetrahedron: Asymmetry 16 (2005) 751–754
11. Hirao, A.; Mochizuki, H.; Nakahama, S.; Yamazaki, N. J.
Org. Chem. 1979, 44, 1720–1722.
14. Brown, H. C.; Garg, C. P. J. Am. Chem. Soc. 1961, 83,
2952–2953.
12. Morrison, J. D.; Grandbois, E. R.; Howard, S. I. J. Org.
Chem. 1980, 45, 4229–4231.
15. Beretta, E.; Cinquini, M.; Colonna, S.; Formasier, R.
Synthesis 1974, 425–426.
´
13. Carter, M. B.; Schiøtt, B.; Gutierrez, A.; Buchwald, S. L.
J. Am. Chem. Soc. 1994, 116, 11667–11670.
16. Brown, C. A.; Krishnamurthy, S. J. Org. Chem. 1978, 43,
2731–2732.