Models and objective functions for the optimisation of selectivity in reversed-phase liquid chromatography.

Interpretive methodologies are the most efficient tools for finding the optimal conditions in chromatography. These methodologies are supported by models or algorithms able to infer the system behaviour upon changes in the experimental factors. Once the models are built with data obtained from sets of carefully designed experiments, molecular modelling or other approaches, they can be applied to predict the performance of new conditions. The different elements involved in these methodologies, for both isocratic and gradient elution, are given. Special attention is devoted to the description of retention, owing to its major impact on the prediction of chromatographic resolution. Several models considering the main factors affecting retention (i.e. organic modifiers, pH and temperature), and procedures that enhance the predictions, are presented. Both the existence of skewed peaks and the effect of elution conditions on peak profiles are considered. Finally, the assessment of resolution, as well as other secondary aims that affect the practical suitability of the optimal conditions, is discussed.