CRITERIA FOR OPTIMIZING THE SEPARATION OF TARGET ANALYTES IN COMPLEX CHROMATOGRAMS

Optimization procedures require adequate criteria to assess the qualit y of the separation. So far, the vast majority of reports on the subje ct deal with the situation in which all peaks need to be separated ('c omplete optimization'). In practice one of the most important requirem ents for optimization criteria is their ability to deal with limited o ptimization, i.e. with situations in which only a limited number of so lutes in a sample is relevant. In this paper, the adaptation of the mo st widely used optimization criteria to limited optimization is discus sed. Among the most useful criteria are the minimum resolution and the calibrated normalized resolution product. The characteristics and obj ectives of these two criteria are discussed in detail. The case of non -ideal separations (featuring asymmetrical peaks and/or peaks of vastl y different areas) is also investigated. The discussion and evaluation of optimization criteria are based on simulated and experimental data obtained during HPLC optimization studies.