A heteroleptic dirhodium
paddlewheel complex comprising three chiral
carboxylate ligands and one achiral acetamidate ligand has recently
been found to be uniquely effective in catalyzing the asymmetric cyclopropanation
of olefins with α-stannylated (silylated and germylated) α-diazoacetate
derivatives. A number of control experiments in combination with detailed
computational studies provide compelling evidence that an interligand
hydrogen bond between the −NH group of the amidate and the
ester carbonyl group of the reactive rhodium carbene intermediate
plays a quintessential role in the stereodetermining transition state.
The penalty for distorting this array outweighs steric arguments and
renders two of the four conceivable transitions states unviable. Based
on this mechanistic insight, the design of the parent catalyst is
revisited herein: placement of appropriate peripheral substituents
allows high levels of diastereocontrol to be imposed upon cyclopropanation,
which the original catalyst lacks. Because the new complexes allow
either trans- or cis-configured stannylated cyclopropanes to be made
selectively and in excellent optical purity, this transformation also
marks a rare case of diastereodivergent asymmetric catalysis. The
products are amenable to stereospecific cross coupling with aryl halides
or alkenyl triflates; these transformations appear to be the first
examples of the formation of stereogenic quaternary carbon centers
by the Stille reaction; carbonylative coupling is also achieved. Moreover,
tin/lithium exchange affords chiral lithium enolates, which can be
intercepted with a variety of electrophilic partners. The virtues
and inherent flexibility of this new methodology are illustrated by
an efficient synthesis of two salinilactones, extremely scarce bacterial
metabolites with signaling function involved in the self-regulatory
growth inhibition of the producing strain.