RESUMO
Chiralpak ZWIX(+) and ZWIX(-), are brush-type bonded-silica chiral stationary phases (CSPs), based on complex diastereomeric Cinchona alkaloids derivatives bearing both a positive and a negative charge. In the present study, we aimed to improve the understanding of retention and enantioseparation mechanisms of these CSPs employed in supercritical fluid chromatography (SFC). For this purpose, 9 other stationary phases were used as comparison systems: two of them are commercially available and bear only a positive charge (Chiralpak QN-AX and QD-AX) and the 7 others were designed purposely to be structurally similar to the parent ZWIX phases, but miss some portion of the complex ligand. First, cluster analysis was employed to identify similar and dissimilar behavior among the 11 stationary phases, where ionic interactions appeared to dominate the observed differences. Secondly, the stationary phases were characterized with linear solvation energy relationships (LSER) based on the SFC analysis of 161 achiral analytes and a modified version of the solvation parameter model to include ionic interactions. This served to compare the interaction capabilities for the 11 stationary phases and showed in particular the contribution of attractive and repulsive ionic interactions. Then the ZWIX phases were characterized for their enantioseparation capabilities with a set of 58 racemic probes. Discriminant analysis was applied to explore the molecular structural features that are useful to successful enantioseparation on the ZWIX phases. In particular, it appeared that the presence of positive charges in the analyte is causing increased retention but is not necessarily a favorable feature to enantiorecognition. On the opposite, the presence of negative charges in the analyte favors early elution and enantiorecognition. Finally, a smaller set of 30 pairs of enantiomers, selected by their structural diversity and different enantioseparation values on the ZWIX phases, were analyzed on all chiral phases to observe the contribution of each structural fragment of the chiral ligand on enantioselectivity. Molecular modelling of the ligands also helped in understanding the three-dimensional arrangement of each ligand, notably the intra-molecular hydrogen bonding or the possible contribution of ionic interactions. In the end, each structural element in the ZWIX phases appeared to be a significant contributor to successful enantioresolution, whether they contribute as direct interaction groups (ion-exchange functions) or as steric constraints to orientate the interacting groups towards the analytes.
