Is the solvation parameter model or its adaptations adequate to account for ionic interactions when characterizing stationary phases for drug impurity profiling with supercritical fluid chromatography?

Nine commercially available polar and aromatic stationary phases were characterized under supercritical fluid chromatographic (SFC) conditions. Retention data of 64 pharmaceutical compounds was acquired to generate models based on the linear solvation energy relationship (LSER) approach. Previously, adaptation of the LSER model was done in liquid chromatography by the addition of two solute descriptors to describe the influence of positive (D(+)) and negative (D(-)) charges on the retention of ionized compounds. In this study, the LSER models, with and without the ionization terms for acidic and basic solutes, were compared. The improved fits obtained for the modified models support inclusion of the D(+) and D(-) terms for pharmaceutical compounds. Moreover, the statistical significance of the new terms in the models indicates the importance of ionic interactions in the retention of pharmaceutical compounds in SFC. However, unlike characterization through the retention profiles, characterization of the stationary phases by modelling never explains the retention variance completely and thus seems less appropriate.