isomers was obtained, indicative of asymmetric induction
by means of carbocations (entries 8 and 9 in Table 2). When
a mixture of 1f and 1e (1e,f) was treated with 2d under the
given conditions, 1f (more hindered than 1e) led to the
coupled product 5e. Similarly, 1h,i (a mixture of 1h and 1i)
afforded 5g in which 1i (more hindered than 1h) was
embedded (entries 5 and 7 in Table 2).
Table 3 summarizes the reactions of cyclic substrates 2f-
h, and again the same regioselectivity as that for the acyclic
versions was observed without fail. Most striking in this
series is that the ring-to-ring coupling reactions could be
achieved so easily that functionalized ring-assemblies such
as 6f-h (entries 5-7, Table 3) were obtained in a rapid and
simple manner: the construction of complex molecules of
this class should be highly beneficial to organic synthesis
and are otherwise difficult to access.
Entries 5 and 7 in Table 3 show that the same events as
those in entry 2 in Table 1 or entries 5 and 7 in Table 2
occurred again. Thus, the more substituted sp2-carbon of enol
silyl ethers has higher reactivity toward the cationic species
in general. These cases efficiently provide quaternary carbon
centers. Clarification of the exact reason responsible for the
unique chemoselectivity8 of this class must await future
theoretical and/or experimental studies.
In conclusion, we have demonstrated highly effective
coupling reactions between enol silyl ethers and carbocation
species that deserve consideration as a practical method for
carbon-carbon bond formation. Substrates 2a-c or 2d-h
can easily be prepared from ketones or enones, respectively,
with corresponding acetylide anions followed by O-silyl-
ation.5 This requires no special skill, conditions, or reagents.
Also, enol silyl ethers 1a-k were available through tradi-
tional protocol uneventfully.4 Thus, the present method is
highly convenient and useful for the introduction of allenic
or enyne functionalities to the R-position of substituted
ketones. Further explorations of this chemistry, including
complex natural product syntheses, are under investigation.
(9) General procedure: to a solution of 1 (0.5 mmol) and 2 (0.6 mmol)
in CH2Cl2 (5 mL) was added Lewis acid (0.05 mmol) at -30 °C. The
reaction was stirred at -30 °C until the disappearance of 1 was indicated
by TLC. To the mixture was added water (20 mL), and the resulting mixture
was extracted several times with 1:3 ethyl acetate-hexane. The combined
organic layers were dried over anhydrous Na2SO4 and concentrated to give
a crude residue, which was purified by column chromatography on silica
gel using ethyl acetate-hexane to afford 4, 5, or 6. For the reaction of 1a
with 2a, TMSOTf afforded 60% yield of 3a (Scheme 1) and BF3‚OEt2
afforded 80% yield (Table 1). Therefore, in this work, BF3‚OEt2 was used
as the Lewis acid except for the case of entry 4 in Table 2, where TMSOTf
was used.
Acknowledgment. This research was supported by a
Grant-in-Aid for Scientific Research from the Ministry of
Education, Culture, Sports, Science and Technology, Japan.
(10) The reason for the preferential formation of (Z)-enynes was already
discussed; see ref 3.
(11) Structures and isomer ratios for the products were determined by
careful analysis of NMR spectra, although the relative configurations of
major isomers were not determined yet; see also Supporting Information.
(12) Reactions of 1l with 2a,d, 2i with 1a, and 2j with 1a,d did not take
place under the given reaction conditions.
Supporting Information Available: Experimental pro-
cedures and spectroscopic data for coupling products 3-6
1
and copies of H and 13C NMR spectra for representative
compounds. This material is available free of charge via the
(13) Denmark, S. E.; Pham, S. M. Org. Lett. 2001, 3, 2201-2204.
OL0271537
54
Org. Lett., Vol. 5, No. 1, 2003