electrophile.11 Various conditions and catalysts were then
tested (Table 1). BINOL-derived phosphoric acid rac-3a was
rapidly identified as a satisfying catalyst candidate when
excess ketone was used (Table 1, entry 5).
The reactions carried out under equivalent molar amounts
of reagents or excess of glyoxylate afforded the aldol product
in slightly lower yields. Surprisingly, concerning the required
acidity of the catalyst, stronger acids such as 1 and 2 behaved
sluggishly (Table 1, entries 3 and 4).
In order to induce high selectivity in asymmetric transforma-
tions when using BINOL-derived phosphoric acids, the chiral
BINOL backbone must be substituted by bulky aromatics on
the 3,3′ positions. Several 3,3′-disubstituted H8-BINOL-derived
phosphoric acids were thus prepared according to an improved
procedure developed in our laboratory.12
The introduction of aromatic groups with variable steric
hindrance had a strong influence on both diastereo- and
enantioselectivity. Phosphoric acid 4a with a phenyl sub-
stituent afforded almost no diastereoselectivity and an
average enantioselectivity (er 75:25) for the syn diastereoi-
somer (Table 1, entry 8), whereas hindered acids 3c13 (Ar
) 2,4,6-(i-Pr)3C6H2) and 4e (Ar ) 2,4,6-(Me)3C6H2) led,
respectively, to improved diastereoselectivity in favor of the
syn isomer (dr syn/anti 80:20, Table 1, entries 7) and
enantioselectivity (er 86:14) for the major syn isomer (Table
1, entry 12).
Figure 1. Phosphoric acid catalyisis versus enamine catalysis.
involving chiral zinc or rhodium complexes describe the
successful use of methyl vinyl ketone and cyclic enones in
asymmetric direct aldol reactions, thus illustrating the
challenge represented by those substrates.5,6
With this background in mind, we became interested in
the alternative well-known acid catalysis. Chiral Lewis acids
have been frequently used to control the stereoselectivity of
the direct aldol reaction.7 Witnessing the rapid growth of
the use of chiral Brønsted acids, we embarked in an
evaluation of those catalysts for the aldol reaction.8 To the
best of our knowledge, among the rare cases of carbonyl
group activation9 such a strategy has been ignored so far.10
Next, we screened the reaction conditions to optimize the
yield and selectivity of the aldol product. Among the various
nonprotic solvents tested (toluene, xylene, THF, Et2O, Bu2O,
CH2Cl2, CH3CN), toluene was identified to afford the highest
diastereo- and enantioselectivity at room temperature (see
the Supporting Information). Interestingly, reactions could
be performed at 50 °C with little erosion of enantioselectivity
(data not shown). Decreasing the reaction temperature had a
beneficial impact on the overall selectivity (Table 1, entries 13
and 16). An optimal amount of 5 mol % catalyst was
determined, while 1.5 mol % led to slightly decreased selectivity
and 10 mol % to no improvement (data not shown). Carrying
out the reaction in nearly solventless conditions, using only the
toluene provided by the commercial solution of the glyoxylate,
led to a significant improvement of both reaction rates and
selectivities (dr syn/anti 70:30 and 93:7 er for the syn isomer,
Table 1, entry 14).
Initially we screened various Brønsted acids and substrates
to identify suitable conditions to further investigate this
strategy. Cyclohexanone was selected as a model nucleophile,
while ethyl glyoxylate was found to be the most reactive
(5) Trost, B. M.; Shin, S.; Sclafani, J. A. J. Am. Chem. Soc. 2005, 127,
8602–8603
.
(6) Mizuno, M.; Inoue, H.; Naito, T.; Zhou, L.; Nishiyama, H.
Chem.sEur. J. 2009, 15, 8985–8988
.
(7) (a) Trost, B. M.; Silcoff, E. R.; Ito, H. Org. Lett. 2001, 3, 2497–
2500. (b) Kumagai, N.; Matsunaga, S.; Kinoshita, T.; Harada, S.; Okada,
S.; Sakamoto, S.; Yamaguchi, K.; Shibasaki, M. J. Am. Chem. Soc. 2003,
125, 2169–2178.
(8) (a) Akiyama, T. Chem. ReV. 2007, 107, 5744–5758. (b) Terada, M.
Chem. Commun. 2008, 35, 4097–4112. (c) Terada, M. Synthesis 2010, 1929–
1982.
(9) (a) Xu, S.; Wang, Z.; Wang, X.; Zhang, X.; Ding, K. Angew. Chem.,
Int. Ed. 2008, 47, 2840–2843. (b) Rueping, M.; Nachtsheim, B. J.; Moreth,
S. A.; Bolte, M. Angew. Chem., Int. Ed. 2008, 47, 593–596. (c) Rowland,
E. B.; Rowland, G. B.; Rivera-Otero, E.; Antilla, J. C. J. Am. Chem. Soc.
2007, 129, 12084–12085. (d) Rueping, M.; Ieawsuwan, W.; Antonchick,
A. P.; Nachtsheim, B. J. Angew. Chem., Int. Ed. 2007, 46, 2097–2100. (e)
Nakashima, D.; Yamamoto, H. J. Am. Chem. Soc. 2006, 128, 9626–9627.
(f) Rueping, M.; Theissmann, T.; Kuenkel, A.; Koenigs, R. M. Angew.
Chem., Int. Ed. 2008, 47, 6798–6801. (g) Akiyama, T.; Katoh, T.; Mori,
K. Angew. Chem., Int. Ed. 2009, 48, 4226–4228. (h) Lu, M.; Zhu, D.; Lu,
Y.; Zeng, X.; Tan, B.; Xu, Z.; Zhong, G. J. Am. Chem. Soc. 2009, 131,
4562–4563. (i) Gu, Q.; Rong, Z.-Q.; Zheng, C.; You, S.-L. J. Am. Chem.
Soc. 2010, 132, 4056–4057. (j) Sun, F.-L.; Zeng, M.; Gu, Q.; You, S.-L.
Chem.sEur. J. 2009, 15, 8709–8712.
Recycled catalyst, easily recovered by column chroma-
tography in 80% yield, performed with the same level of
selectivity (Table 1, entry 15). The absolute configuration
of the main isomer was acertained by comparison with
literature data.14
Interestingly, the observed diastereoselectivites were mod-
erated in favor of the syn diastereoisomer, so we decided to
(10) For previous asymetric Mukaiyama aldol reactions catalyzed with
chiral Brønsted acids, see: (a) Cheon, C. H.; Yamamoto, H. Org. Lett. 2010,
12, 2476–2479. (b) Garc´ıa-Garc´ıa, P.; Lay, F.; Garc´ıa-Garc´ıa, P.; Rabalakos,
C.; List, B. Angew. Chem., Int. Ed. 2009, 48, 4363–4366. (c) McGilvra,
J. D.; Unni, A. K.; Modi, K.; Rawal, V. H. Angew. Chem., Int. Ed. 2006,
45, 6130–6133. (d) Gondi, V. B.; Hagihara, K.; Rawal, V. H. Angew. Chem.,
Int. Ed. 2009, 48, 776–779. (e) Zhuang, W.; Poulsen, T. B.; Jørgensen,
K. A. Org Biomol. Chem. 2005, 3, 3284–3289. For previous activation of
ethyl glyoxylate by chiral phosphoric acids, see: (f) Terada, M.; Soga, K.;
Momiyama, N. Angew. Chem., Int. Ed. 2008, 47, 4122–4125. (g)
Momiyama, N.; Tabuse, H.; Terada, M. J. Am. Chem. Soc. 2009, 131,
12889–12890.
(11) Other electrophiles were tested (2-methylpropanal, 1-butanal,
chloral, 2- and 3-pyridine carboxaldehyde, benzaldehyde, phenyl glyoxal),
and only traces of product were detected. Ethyl trifluoropyruvate delivered
42% yield.
(12) Pousse, G.; Devineau, A.; Dalla, V.; Humphreys, L.; Lasne, M.-
C.; Rouden, J.; Blanchet, J. Tetrahedron 2009, 65, 10617–10622.
(13) During the revision process of the manuscript, the octahydrogenated
analogue 4f (R ) 2,4,6-(i-Pr)3C6H2) was prepared and the obtained
selectivities were identical to those of 3c (at 0 °C: 75% yield, 80:20 dr,
89:11 er).
Org. Lett., Vol. 12, No. 16, 2010
3583