Since eutomer and distomer show different biological effects, access to pure enantiomers is essential. Preferential enrichment is a relatively new process of chiral resolution developed in the 90s. Extension of its scope was attempted with tryptophan ethyl ester hydrochloride. This salt meets one of the most important requirements for preferential enrichment, namely, a larger solubility for the pure enantiomers with reference to the racemic compound. But, due to the absence of a solid solution, this salt cannot perform preferential enrichment. This impossibility was endorsed by semi-empirical molecular modeling.

Preferential enrichment (PE) is a crystallization process, starting from either a racemic of slightly enantio-enriched solution (ca. +5%) that results in a high enantiomeric excess in the liquid phase (>+90%ee) and a slight opposite excess in the deposited crystals (−2 to −5%ee). The mechanism(s) of this symmetry-breaking phenomenon is (are) still a matter of debate since it eludes rationalization by phase diagram formalism. In this publication, we thoroughly reinvestigate the PE phenomenon of arginine fumarate by using a new approach: the process is monitored by introducing isotopically labeled arginine enantiomers into the crystallization medium to better understand the mass exchanges during crystallization. These experiments are supported by chiral HPLC-MS/MS. This study permits re-evaluating the criteria that were thought mandatory to perform PE. In particular, we show that PE occurs by a continuous exchange between the solution and the crystals and does not require the occurrence of a solvent-mediated solid–solid phase transition.