Credit: © 2008 NPG

Nature has extraordinary specificity in the recognition of chiral molecules, thus the ability to synthesize a single enantiomer is an important goal. In some cases, methods are well developed, but there remain several classes of molecules where the problem remains far from solved. Of particular note are methods involving the construction of quaternary stereocentres, such as those in chiral tertiary alcohols. Chiral secondary alcohols can be produced enatioselectively with relative ease by the addition of a carbon nucleophile to an aldehyde. However, such methods are not amenable to the production of chiral tertiary alcohols, because the selectivity usually relies on the steric difference between the substituents flanking the carbonyl group — the steric difference is much smaller in the necessary ketone starting compound.

Now, Varinder Aggarwal and co-workers from the University of Bristol have developed1 a simple method to convert a chiral secondary alcohol to either enantiomer of a tertiary alcohol. The method relies on an umpolung (reversed polarity) disconnection of the desired carbon–carbon bond. The secondary alcohol is converted first to a carbamate, rendering the methine proton just acidic enough to be removed with a strong lithium base. The resulting organolithium species reacts with either a boronic ester or a borane, and a rearrangement followed by an oxidative workup generates the desired tertiary alcohol.

Crucially, the use of the borane results in the stereochemistry being almost completely inverted, whereas the boronic ester results in it being retained. Thus, either enantiomer of a tertiary alcohol can be produced from just one enantiomer of the starting material simply by the choice of an achiral reagent.