~) Pergamon
Tetrahedron Letters, Vol. 37. No. 47, pp. 8471-8474, 1996
Copyright © 1996 Elsevier Science Ltd
Printed in Great Britain. All rights reserved
PII: S0040-4039(96)01969-7 o040-4039/96 $15.00 ÷ 0.00
An ab initio Molec~d~r Orbital Study on the Lewis Acidity of TMS-CI and
TMS-CN Toward an a,~-Unsaturated Aldehyde: Are These Acid-Base
Interactions Important in Organocuprate 1,4-Additions to Enones?
Bruce H. Lipshutz,* Donald H. Aue,* and Brian James
Department of Chemistry
University of California, Santa Barbara, CA 93106
[Fax: 805-893-4120]
Abstract: High-level ab initio calculations were used to determine the effectiveness of commonly used
silanes as Lewis acids toward acrolein as a model for Gilman cuprate co~ugate addition reactions.
The data suggest that complexation with an enone is by a weak dipole.dipole interaction and is not
directly responsible for the rate accelerating effect of TMS-X Copyright © 1996Elsevier Scicncc Ltd
The remarkable effect of TMS-CI as an additive in conjugate addition reactions of cuprates
R2CuLi has previously been attributed, in part, to its role as a Lewis acid. I Recently, TMS-CN has
been shown to elicitthe same type of response; that is, to increase both rates and yields of 1,4-
additions to enals/enones. 2 While there is little argument regarding the potential for BF3 to associate
with a carbenyl oxygen lone pair, it is far less obvious that a related acid-base interaction exists to
any significant extent with either TMS-CI or TMS-CN. These two silanes are unique relative to
other silylderivatives,e~g., TMS-Br, TMS-I, and TMS-OTfin that at low temperatures (between -50
and -78°C) they do not react rapidly with Gi]msrt cuprates to form R-TMS and LiCI. s They do,
however, based on recently disclosed multinuclear NMR spectral evidence, appear to associate to at
least some extent with R2CuLi, where the chlorine in TMS-CI and the nitrilegroup in TMS-CN act as
Lewis bas/c ligands toward lithium~3 or copper within the cluster. 4 Such interactions, nonetheless,
do not exclude the possibilitythat a purely Lewis acid interaction (between enone and TMS-X) could
be operating as well. Although there is no physical data~ (e.g., IR or NMR) supporting the notion that
either TMS-CI or TMS-CN acts in this capacity toward an aldehyde or ketene carbonyl, high level ab
initio calculations 8 were undertaken to assess the extent to which such interactions might exist.
Considering the growing number of TMS-X influenced processes in synthetic chemistry, 7 and the
postulation of Lewis acid complexation to explain the effects of TMS-X in these processes, l's it was
felt that such a study might have far broader implications than just those related to cuprate Michael
additions. We now describe the preliminary results of our calculations.
Acrolein (1) was chosen as our model ct,~-unsaturated carbonyl compound. The presumed
Lewis acids chosen for study were cldorosilane (C1-SiH3, 2), cyanosilane (H3Si-CN, 3), TMS-C1 (4),
8471