Organic Syntheses, Vol. 79, pp. 116-124
Checked by Eric M. Flamme and William R. Roush.
1. Procedure
2. Notes
1.
(1R,2S)-(+)-1 was
prepared according to the accompanying procedure (Abiko, A.
Org. Synth.
2002,
79, 109).
3. A stock solution
(1 M) of dicyclohexylboron
trifluromethanesulfonate was prepared according to the accompanying
procedure (Abiko, A.
Org. Synth.
2002,
79, 103). Two equivalents
of the boron triflate are necessary for complete enolization of the ester. When one
equivalent is used, the enolization proceeds only to 50% conversion.
5. The reaction may be monitored by TLC using a solvent mixture of
90% hexanes and 10%
ethyl acetate. Starting material
1 has an R
f
= 0.37 and the product
2 has an R
f = 0.25. Both spots were visualized
by UV detection. It is possible to visualize the minor diastereoisomer by using a
dilute TLC sample and a solvent system of
85%
hexanes and 15% ethyl acetate.
Under these conditions product
2 has an R
f = 0.32 and the minor
aldol diastereoisomer has an R
f = 0.27.
6. HPLC analysis [
21 mm Dynamax-60A column
(Si 83-111-C),
78% hexanes-22%
ethyl acetate, 10 ml/min flow rate, UV detection;
retention time for
1 = 14 min; retention time for
2 = 16.5 min; retention
time for minor aldol diastereoisomer = 17.9 min] of the crude reaction product showed
that the residue contained 10-15% of the starting ester and that the two anti aldol
products were obtained in 96 : 4 ratio. The submitter found the product ratio to be
97 : 3 by HPLC analysis, with less than 3% of the starting ester remaining.
7. Attempts to further purify this material by additional recrystallizations
were unsuccessful. The submitter indicated that diastereomerically pure
2 could
be obtained by additional recrystallizations from
85 : 15
hexanes-ethyl acetate, but the checkers were unable
to achieve this result. However, the two aldol diastereomers can be separated chromatographically,
as described in the body of the procedure.
8. The submitter performed this chromatography step using a 5 : 1
mixture of
hexane and ethyl
acetate. However, the checkers found that the mother liquors
did not dissolve in this solvent mixture, and added
dichloromethane
to the chromatography mixture to solve this problem.
9. The physical and spectral data of (+)-
2 are as follows:
mp 142-142.5°C,
[α]D23 19.7° (c 2.05, CHCl3);
1H NMR (CDCl
3) δ:
0.90 (d, 3H, J = 6.7), 0.95 (d, 3H, J = 6.8), 1.10
(d, 3H, J = 7.2), 1.17 (d, 3H, J = 7.0), 1.73, (m,
1H), 2.28 (s, 3H), 2.37 (br s, 1H, OH), 2.49 (s, 6H),
2.62 (dq, 1H, J = 7.1, 7.2), 3.41 (br, 1H), 4.11 (dq,
1H, J = 4.4, 7.0), 4.55 (1H, A of ABq, J
AB = 16.5),
4.79 (1H, B of ABq, J
AB = 16.5), 5.82 (d, 1H, J = 4.4),
6.82-6.86 (m, 2H), 6.87 (s, 2H), 7.12-7.33 (m, 8H);
13C NMR (CDCl
3) δ:
13.4, 14.2, 15.5, 19.8, 20.7,
22.8, 30.0, 42.9, 48.1, 56.7,
77.6, 78.1, 125.8, 127.0, 127.6,
127.8, 128.2, 128.3, 132.0, 133.3,
138.1, 138.5, 140.1, 142.4, 174.8.
Anal. Calcd for C
32H
41NO
5S: C, 69.66; H, 7.49; N,
2.54. Found: C, 69.84; H, 7.62; N, 2.53.
10. In one run (5.0-mmol scale), the checkers obtained a crude product
that contained ca. 40% of recovered
1. Attempts to purify the aldol product
from this mixture by using the described crystallization procedure was unsuccessful.
Accordingly, the crude product (2.36 g) was purified by flash chromatography on
170 g of silica gel using
1000
ml of a 9 : 1 : 1 mixture of hexane, ethyl
acetate and methylene chloride. This provided
0.72 g of starting ester
1,
1.39 g
of pure aldol
2, and
0.259 g
of mixed fractions containing
2 and the minor aldol diastereoisomer (4 : 1
by 1H NMR analysis).
11. If the auxiliary does not precipitate during the trituration
step, the solution is not sufficiently dry or may contain too much
dichloromethane.
Under these circumstances, the solution should be concentrated, redried with
sodium
sulfate, and the trituration procedure repeated. The checkers also found
that trituration is best performed by spinning the flask on a
rotary evaporator
(at atmospheric pressure) for several minutes. Swirling the flask by hand was not
always successful.
12. The checkers found that small amounts of ester
2 (2-7%)
remained unreacted, even when
1.1 equiv of lithium
aluminum hydride was used for the reduction. Progress of the
reduction can be monitored by TLC (
70%
hexane, 30% ethyl acetate):
2, R
f = 0.75;
3, Rf = 0.2;
4, R
f = 0.75.
Because
2 and
4 co-elute under the chromatography conditions, auxiliary
4 recovered by chromatography is not pure.
13. The submitter and checkers have successfully performed this procedure
on double the reaction scale [10 mmol of (+)-
2], and obtained the product in
92% yield.
14. The spectroscopic properties of product (2S, 3R)-
3 are
as follows:
[α]D23
19.6° (c 0.57, CHCl3), lit.
219.6° (c 0.75, CHCl3);
1H NMR (CDCl
3) δ: 0.78
(d, 3 H, J = 7), 0.81 (d, 3 H, J = 6.6), 0.87 (d, 3 H, J
= 7.0), 1.73 (m, 2H), 3.24(dd, 1 H, J = 3.4, 8.0),
3.53 (1 H, B of ABX, J
AB =10.8, J
BX = 7), 3.65
(1 H, A of ABX, J
AB = 10.8, J
AX = 3.7), 3.70 (br
s, 2 H, OH);
13C
NMR (CDCl
3) δ: 14.0, 15.2, 20.0,
30.5, 37.2, 68.0, 81.8;
Anal. Calcd for C
7H
16O
2: C, 63.60; H, 12.20. Found:
C, 63.42; H, 12.01.
All toxic materials were disposed of in accordance with "Prudent Practices in the
Laboratory"; National Academy Press; Washington, DC, 1995.
3. Discussion
Several methods for the anti-selective, asymmetric aldol reaction
3 recorded in the literature include (i) the
use of
boron,
titanium, or tin(II) enolate
carrying chiral ligands,
4 (ii) Lewis acid-catalyzed aldol reactions
of a metal enolate of chiral carbonyl compounds,
5 and (iii) the use of the metal
enolate derived from a chiral carbonyl compound.
6 Although many of these methods provide anti-aldols
with high enantioselectivities, these methods are not as convenient or widely applicable
as the method reported here, because of problems associated with the availability
of reagents, the generality of reactions, or the required reaction conditions.
The present procedure is based on the original report by the author and co-workers,
7 and utilizes the characteristic
features of the boron-mediated aldol reaction of carboxylic esters, represented by
the ability to produce either anti- or syn-aldols under the specified reaction conditions.
8
The reliability and practicality of the boron- mediated aldol reaction have been demonstrated
by many examples.
9 Both enantiomers of
the chiral auxiliary alcohol in this procedure are prepared from readily available
(−)- or
(+)-norephedrine in three easy steps in high overall
yield.
10 The auxiliary
alcohol could be recovered in nearly quantitative yield (and reused) with the transformation
of the aldol products to chiral diols or other derivatives. The stoichiometry of the
boron triflate required for the enolization of carboxylic esters was determined empirically.
The present procedure is applicable to a wide range of aldehydes with high selectivity
(both syn:anti and diastereofacial selectivity of anti isomer); see the following
Table.
7 An application to a natural product synthesis
has been reported.
11
Copyright © 1921-2002, Organic Syntheses, Inc. All Rights Reserved