chain, had yet to be determined. Although the Cu(I)
participating in these reductions must be ligated at some point
by one or both phosphorus atoms in 1 or 2, both sulfur and/
or nitrogen (as part of a heteroaromatic array) also form
strong bonds to copper(I) and could sequester the metal,
thereby terminating the catalytic cycle.
Table 2. Effect of THF as a Solvent on Product ee
The study began with three acetylpyridines 3-5, as
summarized in Table 1 (entries 1-3). In all cases, 0.05 mol
entry
reagent
solvent(s)
product ee (%)
%
nonracemic biaryl bisphosphine ligand was chosen as a
representative of the high substrate to ligand (S/L) ratios
possible (thus, in these examples, S/L ) 2000:1). Neither
1
2
3
4
5
6
7
2a
1b
1b
1b
1a
1a
1a
100% toluene
100% toluene
12.5% THF/tol
100% THF
100% toluene
12.5% THF/tol
100% THF
91
90
88
70
78
74
67
2- nor 3-acetylpyridine could be run in the preferred solvent
toluene at -78 °C (0.5 M) due to limited solubility at this
temperature. The former reacted completely at -50 °C using
the Takasago SEGPHOS ligand-based system 2a [(-)-
DTBM-SEGPHOS]10 to afford the (R)-product alcohol in
high yield and with an ee of 90%. By way of comparison,
although reactions of 3 under the influence of the corre-
sponding reagents 1a and 1b proceeded to a limited extent
at -78 °C, the ees in each case were actually lower than
that realized with reagent 2a at the higher reaction temper-
ature. Likewise, 4 required warming to -35 °C to reach
completion after 8 h. Use of THF as a cosolvent with
11
drosilylation. Acetylated thiophene and pyrrole nuclei,
represented by potential educts 11-13, were completely
inert. Remarkably, even upon warming of these three
reactions to ambient temperatures, and independent of ligand,
no reduction took place. Addition of acetophenone to these
reaction mixtures at room-temperature did not result in any
1,2-addition of hydride to this aryl ketone. Why there would
be special opportunities for sequestering copper by sulfur
and nitrogen (assuming this accounts for the lack of reactivity
observed) in these 2-acylated five-membered ring arrays,
when related chelation is possible in heteroaromatics 3, 8,
9, and 10, remains unclear at this time. Attempts to alter the
substrate and/or reagent by introduction of a Lewis acid (e.g.,
4-acetylpyridine 5 allowed for complete reaction at -78 °C
(entry 3), although some erosion in ee was observed. This
pattern due to solvent (vide infra) was unexpected on the
7
basis of prior observations involving aromatic ketones,
where reactions run solely in THF afforded ees essentially
identical to those obtained in 100% toluene.
Two examples in the furan series were examined (Table
12
1
, entries 4 and 5). Again, SEGPHOS-ligated CuH was the
reagent of choice. Although methyl ketone 6 reacted at -50
C to give a product of 92% ee, substrate 7 bearing a larger
BF ‚OEt , (EtO) B) or alternative source of sulfur(II) (e.g.,
dimethyl sulfide; 2 equiv) did not have an impact on reagent
activity.
3
2
3
13
°
side-chain led to less steric discrimination by 2a in the
7
presumed four-centered transition state for hydride delivery.
A brief survey of solvent and ligand effects on the ees of
the furanyl alcohol derived from 6 is illustrated in Table 2.
Clearly, although at -78 °C these reactions did not go to
completion, the percentage of THF present can significantly
alter the extent of stereoinduction. Moreover, under otherwise
identical conditions, while reagents 1b and 2a led to products
of essentially the same ee in pure toluene (entries 1 vs 2),
In summary, an effective method14 has been developed
for the asymmetric 1,2-reduction of heteroaromatic ketones
1a was considerably less effective in this medium (entry 5).
2
-Acetylthiazole 8 could be smoothly reduced at -50 °C
(
12) Noyori, R.; Ohkuma, T.; Koizumi, M.; Yoshida, M. Org. Lett. 2000,
2
, 1749.
in toluene with 2a (Table 1, entry 6). In keeping with the
apparent trend noted above, the BIPHEP-ligated CuH 1b,
even at -78 °C, produced the desired alcohol in lower ee.
The fully substituted 5-acetylthiazole 9 reacted readily at the
same temperature to afford essentially a single enantiomeric
alcohol. Acylated isoxazole 10, sterically akin to furan 7
entry 5), led to a similar rate and level of stereoinduction
entry 8).
Unfortunately, not all of the heteroaromatics investigated
(
(
13) Lipshutz, B. H.; Ung, C.; Sengupta, S. Synlett 1989, 1, 64.
14) Representative Procedure for DTBM-SEGPHOS‚CuH-Cata-
lyzed Asymmetric Hydrosilylations of Heteroaromatic Ketones (Table
, Entry 1). To a flame-dried 25 mL round-bottom flask (RBF) equipped
1
with a magnetic stir bar and purged with argon in a glovebox were added
commercial Cu(I)Cl (4.5 mg, 0.045 mmol, 1 mol %) and NaO-t-Bu (4.4
-
3
mg, 0.045 mmol, 1 mol %), (-)-DTBM-SEGPHOS (2.7 mg, 2.25 × 10
mmol, 0.05 mol %), and 6.0 mL of toluene at room temperature. The mixture
was stirred for 30 min before being cooled to -50 °C. PMHS (1.10 mL,
18.0 mmol, 4 equiv of hydride) was added to the RBF and stirred for 15
min. In a separate 10 mL pear-bottomed flask (PBF) were combined
(
(
2
-acetylpyridine (0.51 mL, 4.50 mmol) and 3.0 mL of toluene under argon,
underwent this copper hydride-catalyzed asymmetric hy-
and the mixture was cooled to -50 °C. The contents of the PBF were
transferred via cannula to the RBF, and the reaction was monitored by TLC.
Upon completion (2 h), the reaction was quenched with 2.5 M aqueous
NaOH (15 mL) and THF (15 mL) and stirred for 3 h. The biphasic mixture
was salted out with NaCl and extracted with Et2O (3 × 10 mL). The
combined extracts were washed with brine and dried over anhydrous
MgSO4. The solvent was removed under reduced pressure to yield 530 mg
(97%) of a colorless oil.1
(10) Saito, T.; Yokoawa, T.; Ishizakik, T.; Moroi, T. Sayo, N.; Miura,
T.; Kumobayashi, H. AdV. Synth. Catal. 2001, 343, 264. The absolute
stereochemistry of (-)-DTBM-SEGPHOS has yet to be established.
(11) Product alcohol from 4 using reagent 2a: (R)-1-(3-pyridinyl)ethanol,
[
R]D +28.2 (c 1.06, MeOH) (lit. [R]D +28.15 (c 1.06, MeOH)); Dictionary
2,15
of Organic Compounds, 6th ed.; Chapman & Hall: London, 1996; p 5525.
Org. Lett., Vol. 4, No. 23, 2002
4047
Lipshutz, Bruce H.
