Scheme 1. Four-Component Direct Aldol Addition Reaction
Table 1. Direct Aldol Addition Reaction with Various
Aldehydes
to generate an R,â-unsaturated thioester (1 f 2), which
would be followed by 1,4-addition of the second thiolate
equivalent to give a thioester enolate (3) and, ultimately, aldol
addition (3 f 4). This chemoselective mode of enolate
formation would preclude aldehyde enolization and, conse-
quently, self-addition. Thus, the need for prior enolate
formation would be eliminated, while maintaining the level
of chemoselectivity associated with such techniques. More-
over, since the cascade sequence is initiated by thiolate
addition, background reactions involving trace amounts of
moisture in the atmosphere or solvent should not be a factor,
and low temperatures would not be required, further sim-
plifying the process. Additionally, the organosulfur aldol
products could participate in numerous subsequent transfor-
mations, leading to an array of useful structures.
To test the feasibility of the proposed four-component aldol
addition reaction, PhSNa (2 equiv) was added to a mixture
of acryloyl chloride (1) (1 equiv) and benzaldehyde (5) (1
equiv) in CH2Cl2.5 However, no aldol adduct was obtained,
and instead, protonated 3 (R ) Ph) was isolated in 92% yield.
Varying the solvent and counterion (Li+, K+) gave no
improvement. We next tried PhSLi in the presence of MgBr2‚
OEt2,2h,i which gave the aldol addition product (11) in 67%
yield within only 30 min. Remarkably, the reaction was
highly selective for the anti diastereomer, which is uncom-
mon in aldol additions,1,6 with an anti-syn ratio of 13:1.
Prolonged reaction time did not improve the yield or affect
the diastereomeric ratio. However, the efficiency was
improved using 3 equiv of PhSLi, 1.5 equiv of 1, 1.2 equiv
of MgBr2‚OEt2, and 1 equiv of 5, which gave 88% yield of
11, with the same anti-syn ratio (Table 1, Entry 1). As
hypothesized, control experiments showed no difference
between anhydrous7 and nonanhydrous8 conditions. We also
ruled out an alternative reaction pathway initiated by thiolate
1,4-addition to 1 to give an acid chloride enolate intermedi-
ate, which would then undergo aldol addition, followed by
Cl f S acyl transfer to give 4. This was done by conducting
the reaction with only 1.5 equiv of PhSLi (equimolar to 1),
along with 5 (1 equiv) and MgBr2‚OEt2 (1.2 equiv), which
gave acrylate thioester 17 in 93% conversion, with <4% of
11.
With simple and efficient conditions established for the
aldol addition with 5, we investigated the reaction scope with
other aldehydes, both with and without R-protons (Table 1).
In all cases, the four-component transformation proceeded
efficiently with short reaction times. No aldehyde self-
addition products were obtained, thus confirming the com-
patibility of the method with enolizable aldehydes. Adding
further to the significance of this result was that, in each
case, the anti product was strongly favored over the more
commonly obtained syn diastereomer.1,6
We next investigated the origin of the anti-selectivity.
Assuming standard models,1 this could originate from either
kinetic addition of the E-(O)-enolate to the aldehyde or from
the relative thermodynamic stability of the anti and syn
products. Several attempts to trap the enolate or kinetic
addition intermediate under a variety of conditions were
unsuccessful. However, we did establish that the reaction is
reversible, suggesting that the diastereoselectivity is thermo-
dynamically controlled. To do this, PhSLi was added to a
mixture of MgBr2‚OEt2, 1, and 5, and after the reaction was
complete (30 min), 4-methylbenzaldehyde was added and
the reaction was continued for 30 min. This gave an
approximately 1:1 mixture of addition products 11 and 4 (R
) Ph, R1 ) 4-MeC6H4), with a 13:1 anti-syn ratio in each
case.
(4) See for example: Kamimura, A.; Omata, Y.; Mitsudera, H.; Kakehi,
A. J. Chem. Soc., Perkin Trans. 1 2000, 4499-4504. Kamimura, A.;
Mitsudera, H.; Asano, S.; Kakehi, A.; Noguchi, M. Chem. Commun. 1998,
1095-1096. Kamimura, A.; Mitsudera, H.; Asano, S.; Kidera, S. Kakehi,
A. J. Org. Chem. 1999, 64, 6353-6360. Ono, M.; Nishimura, K.; Nagaoka,
Y.; Tomioka, K. Tetrahedron Lett. 1999, 40, 1509-1512. Armitage, M.
A.; Lathbury, D. C.; Mitchell, M. B. J. Chem. Soc., Perkin Trans. 1 1994,
1551-1552. Shono, T.; Matsumura, Y.; Kashimura, S.; Hatanaka, K. J.
Am. Chem. Soc. 1979, 101, 4752-4753.
(5) Initial experiments were done using anhydrous conditions.
(6) Pirrung, M. C.; Heathcock, C. H. J. Org. Chem. 1980, 45, 1727-
1728. Corey, E. J.; Kim, S. S. J. Am. Chem. Soc. 1990, 112, 4976-4977.
Abiko, A. Acc. Chem. Res. 2004, 37, 387-395.
(7) Dry CH2Cl2; Ar atmosphere.
(8) Untreated Aldrich ACS-grade CH2Cl2; open to air.
The inherent thermodynamic preference for the anti or syn
addition product with different acrylate derivatives was
examined (Table 2). With the exception of 20, all thioesters
4664
Org. Lett., Vol. 9, No. 22, 2007