Diastereomeric recrystallisation is a method of chiral resolution of enantiomers from a racemic mixture. It differs from asymmetric synthesis, which aims to produce a single enantiomer from the beginning, in that diastereomeric recrystallisation separates two enantiomers that have already mixed into a single solution.
The strategy of diastereomeric recrystallisation involves two steps. The first step is to convert the enantiomers into diastereomers by way of a chemical reaction. A mixture of enantiomers may contain two isomers of a molecule with one chiral center. After adding a second chiral center in a determined location, the two isomers are still different, but they are no longer mirror images of each other; rather, they become diastereomers. In a prototypical example, a mixture of R and S enantiomers with one chiral center would become a mixture of (R,S) and (S,S) diastereomers. (The R-S notation is explained here.)
The second step, once the diastereomers have formed, is to separate them using recrystallisation. This is possible because enantiomers have shared physical properties such as melting point and boiling point, but diastereomers have different chemical properties, so they can be separated like any two different molecules.