Abstract

The asymmetric aldol addition is among the most powerful reactions in synthetic organic chemistry and has been extensively studied over the past 15 years.1 The strategies for reagentcontrolled asymmetric induction fall into three broadly defined classes (Chart 1): (1) asymmetric modification of the enolate with chiral acyl auxiliaries (A), (2) asymmetric modification of the enolate with chiral metalloid auxiliaries (B), and (3) asymmetric modification of the aldehyde with chiral Lewis acids (C). Each of these strategies has yielded spectacular success, and each has unique advantages and disadvantages. The chiral auxiliary approaches are extremely general and give high selectivities by virtue of the highly ordered nature of the transition structures (closed) which results from the structure of R*/L* and the organizational features of the metal M.1a,c,2 Unfortunately, these reactions have yet to be made catalytic. The chiral Lewis acid approach takes advantage of the Mukaiyama aldol reaction3 of enoxysilane derivatives and is demonstrably catalytic and often diastereoand enantioselective. However, these reactions are less general and the selectivity is most likely dominated by van der Waals interactions which guide the matching of enantiotopic faces in open transition states.3,4 We set for ourselves the goal of inventing a new type of aldol addition reaction which involves the ordered preassembly of enolate, aldehyde and chiral agent for maximum asymmetric influence and which would be catalytic in the chiral reagent. The formulation of this concept, Scheme 1, requires a metal enolate moiety capable of expanding its valence by two and a chiral Lewis base G*. The basis of this proposal for ligandpromoted aldehyde additions finds precedent in our recently disclosed asymmetric allylations (crotylations) with allylic trichlorosilanes.5 We wish to report that the corresponding trichlorosilyl enolates are highly reactiVe agents for the aldol reaction and that their additions can be asymmetrically catalyzed by chiral phosphoramides. While substantial literature exists on the generation and reactions of trichlorostannyl6 and trichlorotitanium7 enolates, the chemistry of trichlorosilyl enolates8 is embryonic by comparison.9 For the initial studies, the use of isolated, purified trichlorosilyl enolates was deemed essential, and we followed the general procedure of Baukov and Lutsenko,8a Scheme 2. Thus, from methyl tributylstannylacetate (1)10 we could obtain the trichlorosilyl ketene acetal 28a as a distillable liquid (bp 25 °C/5 mmHg),11 which thermally isomerized to methyl

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call