I have found one example in the literature where the authors claim that a non-racemic mixture of R and S isomers of an organic compound (a convolutamydine) can be purified by letting it crystallise in part. The R and S isomers form 1+1 aggregates (racemic), increasing the ee of the major isomer (still in solution).
Reference: Tetrahedron (2006), 62, 12017-12024.
Could anyone please help me find more examples of this phenomenon, and/or point me towards a review on this topic? I have recently experienced some weird chiral HPLC results and I want to try and make certain that I haven't seen the same phenomenon.