Critical comparison of retention models for optimisation of the separationof anions in ion chromatography I. Non-suppressed anion chromatography using phthalate eluents and three different stationary phases

A series of mathematical models describing analyte retention behaviour in n on-suppressed ion chromatography of anions has been compared in order to as sess their suitability for inclusion in computer optimisation software for determining the optimal eluent composition for a desired separation. The se ries of models comprised the linear solvent strength model (using both the dominant equilibrium approach and the competing ion effective charge approa ch), the dual eluent species model, the Kuwamoto model, the extended dual e luent species model and the multiple species eluent/analyte model, together with a new empirical model, the end points model. An extensive set of expe rimental retention data obtained for 15 anions (acetate, fluoride, iodate, bromate, chloride, nitrite, bromide, chlorate, nitrate, iodide, oxalate, su lfate, sulfite, thiosulfate and phosphate) on three columns (Waters IC Pak A, Hamilton PRP-X100 and Vydac 302 IC) using phthalate eluents of varying c oncentration and pH was used to evaluate the ability of each model to predi ct retention factors. Statistical comparison of the predicted retention fac tors with those obtained experimentally showed that the performance of the theoretical models improved with the complexity of the model, but none of t he theoretical models could give sufficiently reliable prediction of retent ion factors (especially for divalent analyte ions) for the model to be used in optimisation software. However, the empirical end points model (in whic h a linear relationship is assumed between log k' and log [eluent], but the slope of the relationship is determined empirically) gave satisfactory per formance, with correlation coefficients for all analytes of 0.9953, 0.9840 and 0.9919 for the Hamilton PRP-X100, Vydac 302 IC and Waters IC pal; A col umns, respectively. (C) 1998 Elsevier Science B.V. All rights reserved.