Application of a modified linear solvation energy relationship (LSER) model to retention on a butylimidazolium-based column for high performance liquid chromatography

Previously, a new HPLC stationary phase based on n-butylimidazolium bromide was investigated using a linear solvation energy relationship (LSER) to systematically evaluate the intermolecular interactions between 32 test solutes and the stationary phase. The results and further comparisons with conventional reversed phase systems revealed that retention properties are similar to phenyl phases in both methanol/water and acetonitrile/water mixtures. In this work, the LSER model is extended by including the degree of ionization molecular descriptor, D, which takes into account the pKa of ionizable analytes and the pH of the mobile phase. The D molecular descriptor has been further divided into D+ and D− components that separately account for the ionization of basic and acidic solutes, respectively. This is the first study where the ionization terms for weakly acidic solutes and weakly basic solutes have been separated. LSER results obtained with the expanded solute set with and without the inclusion of the D+ and D− solute descriptors were compared. The improved correlation and standard error obtained for the expanded test set in the presence and absence of the D+ and D− descriptors (R2: 0.987 vs 0.846; SE: 0.051 vs 0.163 for 60% MeOH) support inclusion of these additional terms. Further, the coefficients obtained from the multiple linear regression for the expanded test set with the D+ and D− descriptors were more consistent with the coefficients obtained when the test set included just neutral analytes. In addition, the expanded LSER model did a better job of predicting elution order for the ionizable analytes. This work provides further supporting evidence for the multimodal nature of the butylimidazolium stationary phase.