With concomitant removal of the SiMe3 group on oxygen,
monoprotected E-syn-diol 3a was generated in 70% yield
and g95:5 dr, and Mukaiyama aldol product 4a was
produced in 23% yield. The substrate shows selectivity
for Sakurai reactivity probably because the bulky bis(silyl)
moiety shields both sides of the β-position in 2, making the
competitive Mukaiyama aldol reaction unfavorable.
Switching the R group from SiMe3 to the larger SiEt3 or
smaller Me group dramatically reduced the yield, and
switching to Me also reduced the diastereoselectivity
(entries 4 and 5). Switching R to an electron-withdrawing
benzoyl group completely suppressed the Mukaiyama
aldol pathway, but 3c formed with only a moderate yield
and diastereoselectivity (entry 6). More interesting results
were obtained when we examined the effect of geminal
bis(silane) on the reaction. When both Si1 and Si2 were an
SiMe2t-Bu group, neither the Sakurai nor Mukaiyama
aldolreactionoccurred(entry 7). Thisled us tohypothesize
that if the geminal bis(silyl) moiety consisted of a SiMe3
group and a bulkier SiMe2t-Bu or SiMe2Ph group, the more
reactive SiMe3 might be selectively eliminated. To the best of
our knowledge, few studies have addressed this interesting
selectivity issue.8 As expected, the reaction of 2f provided 3e
in 80% yield, with only the SiMe3 eliminated (entry 8).
Moreover, the ratio of 3 to 4, in this case, was higher than
that in entry 1 due to increased hindrance around the
β-position in 2f. Thus, four different selectivities were real-
ized in a single transformation: Sakurai over Mukaiyama
aldol reaction, elimination of SiMe3 over SiMe2Ph, E- over
Z-configuration, and syn- over anti-diastereoselectivity.
The scope of this reaction was then tested using 2f and
various acetals derived from aryl, alkyl, and alkynyl
aldehydes. All reactions proceeded predominantly via the
Sakurai pathway and gave the monoprotected E-syn-diol
(()-3 in acceptable-to-good yields (Table 2), even though
Mukaiyama aldol products were still obtained in some cases
in yields around 10%. In all cases, the SiMe3 was reliably
eliminated, generating SiMe2Ph-substituted E-vinylsilanes
selectively. The diastereoselectivity was excellent in most
reactions, except for the dr of 86:14 for the sterically less
Scheme 1. General Structure of Geminal Bis(silane) (left);
Sakurai vs Mukaiyama Adol Reaction of 3,3-Bis(silyl)
Silyl Enol Ether with Acetal (right)
In previous work, we proposed the most favorable confor-
mation of 22a is that which minimizes allylic strain and non-
bonded interactions, and which also benefits from a double-
hyperconjugation effect between the two CꢀSi bonds and
the enol double bond. Since both the silyl enol ether and
allyl bis(silane) groups in 2 share the same Z-CdC double
bond, the compound was expected to participate in two
competing pathways in a Lewis acid promoted reaction
with an acetal. Such a reaction could either proceed by a
Mukaiyama5 aldol pathway at the β-position to give
aldehyde 4 or undergo umpolung6 and proceed by a
Sakurai7 pathway at the R-position to give monoprotected
diol 3. Here we report detailed studies of this reaction and
observe that up to four different selectivities can be
achieved in a single transformation.
Table 1. Screening of Sakurai Reaction Conditionsa
(5) (a) Mukaiyama, T.; Banno, K.; Narasaka, K. J. Am. Chem. Soc.
1974, 96, 7503. For selected reviews on Mukaiyama aldol and its
variants, see: (b) Mukaiyama, T. Org. React. 1982, 28, 203. (c) Casiraghi,
G.; Zanardi, F.; Appendino, G.; Rassu, G. Chem. Rev. 2000, 100, 1929.
(d) Palomo, C.; Oiarbide, M.; Garcia, J. M. Chem. Soc. Rev. 2004, 33, 65.
(6) For studies on the umpolung of silyl enol ethers, see: (a) Keck,
G. E.; Abbott, D. E.; Wiley, M. R. Tetrahedron Lett. 1978, 28, 139.
(b) Marshall, J. A.; Jablonowski, J. A.; Elliott, L. M. J. Org. Chem. 1995,
a Reaction conditions: 0.15 mmol of 2, 0.18 mmol of p-Cl-PhCH-
(OMe)2, and 0.22 mmol of Lewis acid in CH2Cl2 (3.0 mL) at ꢀ78 °C for
30 min. b R = H in 3a and 3e. c The E-configuration was assigned based
on J3HꢀH vinylic coupling constants ranging from 18 to 20 Hz in 3. The
syn-stereochemistry was determined by NOE experiments on the acet-
onide of desilylated 3a. d Stereochemistry was not determined. e Isolated
yields after purification by silica gel column chromatography. f Ratios
were determined by 1H NMR spectroscopy.
ꢀ ꢀ
60, 2662. (c) Madec, D.; Ferezou, J. P. Tetrahedron Lett. 1997, 38, 6661.
ꢀ
(d) Marko, I. E.; Dumeunier, R.; Leclercq, C.; Leroy, B.; Plancher,
J. M.; Mekhalfia, A.; Bayston, D. J. Synthesis 2002, 7, 958. (e) Roush,
ꢀ
W. R.; Newcom, J. S. Org. Lett. 2002, 4, 4739. (f) Greene, M. A.; Prevost,
M.; Tolopilo, J.; Woerpel, K. A. J. Am. Chem. Soc. 2012, 134, 12482. (g)
Linclau, B.; Cini, E.; Oakes, C. S.; Josse, S.; Light, M.; Ironmonger, V.
Angew. Chem., Int. Ed. 2012, 51, 1232.
(7) (a) Hosomi, A.; Endo, M.; Sakurai, H. Chem. Lett. 1976, 941. For
selected reviews on Sakurai allylation, see: (b) Fleming, I. Allylsilanes,
allylstannanes and related systems. In Comprehensive Organic Synthesis;
Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 6,
pp 563ꢀ593. (c) Fleming, I.; Dunogues, J.; Smithers, R. Org. React 1989,
37, 57.
(8) (a) Lautens, M.; Ben, R. N.; Delanghe, P. H. M. Angew. Chem.,
Int. Ed. 1994, 33, 2448. (b) Lautens, M.; Ben, R. N.; Delanghe, P. H. M.
Tetrahedron 1996, 52, 7221.
The reaction was initially examined using global SiMe3-
substituted 2a and p-Cl-PhCH(OMe)2 in CH2Cl2 at
ꢀ78 °C. SnCl4 appeared to be a better Lewis acid than
both TiCl4 and BF3 OEt2 for leading the reaction pre-
3
dominantly along the Sakurai pathway (Table 1, entries 1ꢀ3).
Org. Lett., Vol. 15, No. 5, 2013
1069