Application of self-organizing maps for the classification of chromatographic systems and prediction of values of chromatographic quantities

The separation of a complex mixture of inorganic and organic anions by ion chromatography–capillary electrophoresis using a cationic polymer added to the background electrolyte and indirect UV detection has been studied. The addition of unmodified polymer to an electrolyte suitable for indirect detection resulted in the appearance of a system peak due to the counter-anion on the polymer and while the position of the analytes relative to this system peak could be changed, this was found to be an unacceptable approach for mixtures of large numbers of analytes. Although conversion of the polymer to replace the counter-ion with the indirect UV detection probe ion simplified the system, this approach restricted the flexibility of the system because the probe and polymer concentration were necessarily linked. This limitation could be overcome by selecting the appropriate type of probe ion, with probes having a low ion-exchange selectivity coefficient providing greater retention of analytes than probes with a high ion-exchange selectivity coefficient. Three electrolyte systems with different probes (benzoate, chromate and phthalate) were modelled using a previously derived migration equation and this was used to optimise the electrolyte composition to enable the separation of a mixture of 24 inorganic and organic anions within 7 min. The electrolyte composition was then optimised for the analysis of anions in Bayer liquor with the final separation selectivity being substantially improved for selected key analytes.