Scheme 1. Asymmetric Reduction of R-Amino Acetophenone and Its Derivatives
The “one-pot” reactions were performed by stirring the
amino indanol in a THF solution together with a stoichio-
metric quantity of borate ester. After 0.5 h, 1.0 equiv of
good enantioselectivity, in situ recycling of the catalyst is
viable. (Figure 1).
14
Having established the optimum one-pot procedure, the
BH ‚DMS was added, followed by 1.0 equiv of aceto-
3
B-OMe catalyst was applied to a number of prochiral ketones
phenone. For comparison purposes, each reaction was also
carried out by isolation of the catalyst using the conventional
two-step procedure involving azeotropic distillation with
toluene. Analysis of alcohol 1 was performed using a
(
Table 2). In all cases except entry 3, the selectivities were
comparable or better than those obtained previously with the
B-Me amino-indanol catalyst prepared before use. The low
selectivity obtained with isopropyl and cyclopropyl analogues
1
combination of HPLC and H NMR spectroscopy (Table 1).
(entries 3 and 4) is consistent with those previously reported
The results show that all of the borate esters used in the
one-pot procedure perform as effectively as using the isolated
catalyst. Furthermore, inexpensive trimethyl borate could be
used instead of trimethyl boroxine with comparable ee.
Throughout much of the work that we have carried out
with this catalyst, catalyst loadings of 10 mol % have always
provided the optimum level of enantioselectivity. This is
presumably because at this loading the rate of the catalyzed
reaction is sufficiently faster than the background reaction
6
and illustrates the limitations of this system. R-Halogenated
substrates, which are versatile intermediates for further
synthetic manipulations, were reduced with good enantio-
selectivity (entries 7 and 8), as was the R,R-dibromo analogue
(
entry 9). This latter entry is important, since it is a masked
R-hydroxy aldehyde. Only two aliphatic ketones were used
in this study (entries 10 and 11), both giving moderate ee’s.
This is unusual since these are good substrates for the CBS
catalyst, yet in comparison only gave ee’s of 50% and 42%,
respectively, with the B-Me catalyst from cis-aminoindanol.
with BH
3
‚DMS. At lower loadings the rate of the BH ‚DMS
3
reduction becomes more significant, thus leading to a
R-Amino aromatic ketones are important substrates for
asymmetric reductions because the product amino-alcohols
are found in a wide variety of biologically active substances.
Natural products containing this motif have been isolated
lowering of the selectivity. Rate studies using the traditional
CBS catalyst have recently confirmed this.12 To reduce the
effective loading, an iterative procedure was employed
whereby additional equivalents of BH ‚DMS and aceto-
3
15
16
phenone were added to the reaction vessel after a 0.5 h
period. The aim was to determine whether the catalyst was
still active after the reaction was complete and also establish
the catalyst endurance under repeated reaction conditions.
The results illustrate essentially no loss of activity nor
selectivity even after 12 iterative additions of acetophenone
and borane, giving in this case an effective catalyst loading
from the plants Oxytropis myriophylla and Isodon excisus
3
(14) Typical Procedure. Trimethyl borate (0.02 cm , 0.17 mmol) was
added to a solution of (1R,2S) cis-1-amino-indan-2-ol (0.25 mg, 0.17 mmol)
3
in dry THF (1 cm ), and the mixture was stirred at room temperature under
a nitrogen atmosphere for 30 min. BH ‚DMS (0.11 cm3, 1.83 mmol) was
3
3
added, the reaction left to stir for 30 min, and then acetophenone (0.2 cm ,
.67 mmol) in dry THF (2 cm ) was added via cannula. The reaction was
stirred for 30 min, and then MeOH (5 cm ) added. Water (5 cm ) was added,
the solvent evaporated, and the remaining aqueous layers extracted with
3
1
3
3
1
3
of 0.8 mol %. Thus although there is an optimum catalyst
loading that should be employed for efficient reduction with
3
CH2Cl2 (3 × 10 cm ). The combined organic layers were washed with 1
3
3
M HCl (3 × 10 cm ) and water (3 × 10 cm ) and dried over MgSO4.
Filtration and removal of solvent gave crude material. Isolation by column
chromatography on silica gel gave the pure material.
(
12) Xu, J.; Wei, T.; Zhang, Q. J. Org. Chem. 2003, 68, 10146-10151.
(13) Note that in both cases the selectivities are lower than noted in Table
(15) Kojima, K.; Purevsuren, S.; Narantuya, S.; Tsetsegmaa, S.; Jamy-
ansan, Y.; Kimio, I.; Yukio, O. Sci. Pharm. 2001, 69, 383-388.
(16) Lee, C.; Kim, J.; Lee, H.; Lee, S.; Kho, Y. J. Nat. Prod. 2001, 64,
659-660.
1
or as reported previously. This is simply a consequence of the rate of
addition of ketone and borane, which could not be controlled in this
experiment because of the apparatus employed (Radleys Carousel).
Org. Lett., Vol. 6, No. 16, 2004
2807