A Tandem Non-Aldol Aldol Mukaiyama Aldol Reaction

Article abstract of DOI:10.1021/ol0358760

(Equation presented) A new one-pot tandem aldol process is described in which a secondary epoxy silyl ether is converted into the 1,5-bis-silyloxy-3- alkanone in good yield. Thus, treatment of the epoxy silyl ether 8 with TBSOTf and base affords the silyl

Full text of DOI:10.1021/ol0358760

ORGANIC
LETTERS
2003
Vol. 5, No. 24
4705-4707
A Tandem Non-Aldol Aldol Mukaiyama
Aldol Reaction
Michael E. Jung* and Alexandra van den Heuvel
Department of Chemistry and Biochemistry, UniVersity of California,
Los Angeles, California 90095-1569
jung@chem.ucla.edu
Received September 25, 2003
ABSTRACT
A new one-pot tandem aldol process is described in which a secondary epoxy silyl ether is converted into the 1,5-bis-silyloxy-3-alkanone in
good yield. Thus, treatment of the epoxy silyl ether 8 with TBSOTf and base affords the silyl enol ether 9 via non-aldol aldol rearrangement
and addition of benzaldehyde and TBSOTf gives the ketone 10 with 4:1 syn selectivity. The diastereoselectivity changes to an anti preference
for most aldehydes. This anti selectivity overwhelms the normal Felkin−Ahn preference; namely, the 1,5-anti isomer predominates even when
it is anti-Felkin−Ahn.
Over the past decade, ongoing research on the non-aldol aldol
reaction has been carried out in our group. The non-aldol
aldol reaction is an alternative process to obtain aldol prod-
ucts starting with an epoxy alcohol in the presence of a
hindered amine base and silyl triflates. We have extended
the use of the non-aldol aldol from generating aldehydes to
the selective formation of methyl ketones.1 We have now
successfully shown that one can generate in situ a silyl enol
ether, which in the presence of an achiral aldehyde and a
Lewis acid can undergo a Mukaiyama aldol reaction favoring
the formation of the syn 1,5-diol with benzaldehyde and fav-
oring the anti 1,5-diol with many other aldehydes. Both
Evans2 and Paterson3 showed that the anti 1,5-diols were
favored when one used the boron enolate for the aldol reac-
tions. However, when one takes a closer look at the reported
reactions, the protecting group on the â-alkoxy group
influences greatly the outcome of the reaction and usually
favors the anti compound except when this group is tert-
butyldimethylsilyl (TBS). The choice of the Lewis acid also
influences the stereochemical outcome of the aldol reaction.
The non-aldol aldol rearrangement allows one to prepare
either the syn or anti aldol products starting from the (E)-
or (Z)-allylic alcohol, respectively. The chirality is introduced
using the Sharpless asymmetric epoxidation. The alcohol is
protected with a silyl group and then treated with diisopropyl-
ethylamine (DIPEA) and a trialkylsilyl triflate (R3SiOTf).
Although the reactions can be carried out with any relatively
unhindered silyl triflate, e.g., TMSOTf, TESOTf, or TBSOTf,
the cleanest reaction mixtures were obtained using TBSOTf
and the TBS-protected alcohol.
The (E)-allylic alcohols 1 are easily prepared by Grignard
addition to the commercially available 2-methyl-2-pentenal
in the case of R ) Et. For R ) Pr, the 2-methyl-2-hexenal
was prepared in three steps starting from butanal and the
Wittig reagent, 1-carbethoxyethylidenetriphenylphosphorane
(Scheme 1). Reduction and subsequent oxidation afforded
the desired 2-methyl-2-hexenal in high yield, which was then
reacted with methylmagnesium bromide. A Sharpless asym-
metric epoxidation under kinetic resolution conditions af-
forded the erythro epoxy alcohol 3 as the major product
(along with alcohols 2 and 4). The alcohol 3 was protected
with a TBS group and then submitted to the non-aldol aldol
conditions. Depending on the workup, one obtains good
yields of either the methyl ketone 6 or the silyl enol ether 7
(Scheme 2). Both compounds are presumably formed from
(1) (a) Jung, M. E.; van den Heuvel, A. Tetrahedron Lett. 2002, 43,
8169-72. (b) Jung, M. E.; van den Heuvel, A.; Leach, A. G.; Houk, K. N.
Org. Lett. 2003, 5, 3375-3378.
(2) Evans, D. A.; Coleman, P. J.; Coˆte´ B. J. Org. Chem. 1997, 62, 788-
789.
(3) Paterson, I.; Gibson, K. R.; Oballa, R. M. Tetrahedron Lett. 1996,
37, 8585-8588.
10.1021/ol0358760 CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/23/2003

tunately, under these same reaction conditions, enolizable
aldehydes enolized to afford the silyl enol ether and no longer
were reactive toward the silyl enol ether. We were able to
circumvent this problem using a two-step procedure. The
silyl enol ether 9 was isolated, and then PhMe2SiLi was used
to generate the lithium enolate, which was then reacted with
an aldehyde to afford monoprotected 1,5-diols 11 (Scheme
4 and Table 1).6 The selectivity was again in favor of the
Scheme 1
Scheme 4
Scheme 2
syn product for benzaldehyde and cyclohexenecarboxalde-
hyde, but the reactions with all of the other aldehydes were
Table 1. Formation of 1,5-Diols from 9 with PhMe2SiLi
compound
aldehyde (RCHO)
yield (%)a
anti:syn
11a
11b
11c
11d
11e
iPr-CHO
Pr-CHO
PhCH2CH2CHO
cyclohexenyl-CHO
Ph-CHO
59
44
39
48
21
1.7:1
1.5:1
2:1
1:1.5
1:1.7
the rearrangement of 5 (as shown in A) via intermediate B,
which is deprotonated or hydrolyzed.
Since silyl enol ethers are very versatile substrates in
organic synthesis, we decided to submit compound 7 to a
Mukaiyama aldol reaction. We were able to obtain the de-
sired aldol product 10 favoring the syn 1,5-diol with
benzaldehyde and catalytic amounts of TBSOTf (Scheme
3). Only one case of a Mukaiyama aldol reaction using
a These are the isolated yields of pure products with the remainder being
recovered starting material 9.
slightly selective for the anti 1,5-diols. Although the reactions
proceeded in very low conversion (about 21-59%), the
yields of aldol products were essentially quantitative.
We were able to obtain a one-pot NAA (non-aldol aldol)
Mukaiyama aldol reaction by generating the silyl enol ether
in situ and then adding excess BF3‚OEt2 and aldehydes to
the reaction mixture (Scheme 5 and Table 2).7 The methyl
Scheme 3
Scheme 5
TMSOTf and benzaldehyde has been reported,4 while in all
other cases the aldehyde was activated as an acetal deriva-
tive.5
This reaction worked well with non-enolizable aldehydes
as shown for benzaldehyde and cinnamaldehyde. In the latter
case, the Mukaiyama aldol product was obtained in 52%
yield in favor of the syn compound (4:1 syn:anti). Unfor-
ketone 6′ always was a side product of this reaction. Some
aldehyde was produced via migration of the alkyl group
during the non-aldol aldol reaction.1a The reaction proceeded
(6) Fleming, I.; Roberts, R. S.; Smith, S. C. J. Chem. Soc., Perkin Trans.
1 1998, 1209-1214.
(7) (a) Scheidt, K. A.; Banister, T. D.; Tasaka, A.; Wendt, M. D.; Savall,
B. M.; Fegley, G. J.; Roush, W. R. J. Am. Chem. Soc. 2002, 124, 6981-
6990. (b) Roush, W. R.; Banister, T. D.; Wendt, M. D.; Jablonowski, J. A.;
Scheidt, K. A. J. Org. Chem. 2002, 67, 4275-4283.
(4) Yukozawa, T.; Yamaguchi, M.; Nakai, T.; Ishikawa, N. Nippon
Kagaku Kaishi 1985, 11, 2202-2204.
(5) (a) Noyori, R.; Murata, S.; Suzuki, M. Tetrahedron 1981, 37, 3899-
3910. (b) Noyori, R.; Murata, S.; Suzuki, M. J. Am. Chem. Soc. 1980, 102,
3248-3249.
4706
Org. Lett., Vol. 5, No. 24, 2003

enol ether of acetone 20 was treated with 2 equiv each of
benzaldehyde, TBSOTf, and Hu¨nig’s base, we isolated a 70%
yield of a 2.5-3:1 ratio of the syn 21 and anti 22 double-
Mukaiyama aldol products. The structures of these two
isomers were proven by hydride reduction: the syn isomer
21 gave a 3.5:1 ratio of the two alcohols 23ab, while the
anti isomer 22 gave the sole alcohol 24 (the only possible
product), both in 80% yield. Thus, the syn stereochemistry
is favored even without the additional R-stereocenter in the
silyl ether.
Table 2. Formation of 1,5-Diols from 5′ with BF3-OEt2
compound
aldehyde (RCHO)
yield (%)
anti:syn
12a
12b
12c
12d
12e
12f
PhCHO
C5H11-CHO
iPr-CHO
PhCH2CH2CHO
PhCHdCH-CHO
cyclohexenyl-CHO
42
61
56
45
20
20
1:1.5
2.5:1
1:1
1.75:1
2:1
1.2:1
We propose that the mechanism of the Mukaiyama aldol
with most aldehydes, e.g., hexanal and the optically active
aldehyde 12, proceeds via an open transition state in accord
with the work of Evans3,10 and Paterson4 in the case of boron
enolate where the anti 1,5-diol is favored. The aldol reaction
with benzaldehyde is anomalous, and the formation of the
syn-1,5 diol may be due to some unknown electronic effect.
Work is underway to try to better understand and improve
the diastereoselectivity of this process.
to give mainly the anti product 12 for all aldehydes except
benzaldehyde and cyclohexenecarboxaldehyde (Table 2). The
best results were obtained using hexanal as the aldehyde.
Reaction of 5 or 5′ with the optically active aldehyde 13
gave compounds 14 and 15 as the major products (14, 2:1;
15, 3:1), which indicates that this 1,5 anti stereoinduction
from the chiral enolate is more important than the chirality
of the aldehyde (Felkin-Anh preference) (Scheme 6).
In conclusion, we have shown that we can selectively
obtain anti 1,5-diols in a one-pot reaction starting from an
epoxy alcohol via non-aldol aldol rearrangement and Mu-
kaiyama aldol. The reaction with benzaldehyde is catalytic
and favors the syn 1,5-diol. For enolizable aldehydes, the
reactions require the addition of an excess BF3‚OEt2 and
aldehyde.
Scheme 6
Scheme 8
The stereochemistry was proven for the benzaldehyde and
hexanal derivatives. The stereochemistry of the other prod-
ucts was assigned by the similarity of their NMR spectra to
that of the hexanal product. The ketodiol derived from 16
was selectively reduced using the Et3B‚THF/NaBH4 proce-
dure to afford one major compound 17. The compound was
then subjected to acetalization conditions to afford two syn
acetonides 18 and 19 as major compounds (Scheme 7).8 The
Acknowledgment. We thank the National Institutes of
Health (CA72684) for financial support, the National Science
Foundation for support under equipment grant CHE-
9974928, and Drs. K. N. Houk and A. G. Leach for helpful
discussions.
Scheme 7
Supporting Information Available: Experimental pro-
cedures and spectroscopic data for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL0358760
(8) O’Malley, S. J.; Leighton, J. L. Angew. Chem., Int. Ed. 2001, 40,
2915-2917.
stereochemistry was assigned by NOESY experiments and
13C NMR shifts.9
The following results also support the assignment of syn
stereochemistry for the major diastereomer. When the TBS
(9) Rychnovsky, S. D.; Rogers, B.; Yang, G. J. Org. Chem. 1993, 58,
3511-3515.
(10) (a) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Yang, M. G. J. Am.
Chem. Soc. 1996, 118, 4322-4343 and references therein. (b) Evans, D.
A., Bartroli, J.; Godel, T. Tetrahedron Lett. 1982, 23, 4577-4580.
Org. Lett., Vol. 5, No. 24, 2003
4707