Dispersion in Miceller Electrokinetic Chromatography

Davis, Joe

Abstract

The objective of this study is to determine quantitative theory that explains the dependence of plate number N in micellar electrokinetic chromatography (MEKC) on the parameters, 1) organized-media concentration, 2) organic-modifier concentration, 3) buffer type, 4) buffer concentration, and 5) buffer pH. MEKC is a high-resolution electrokinetic chromatography first reported in 1984 that has been used widely in the analysis of compounds of interest to health-related professions, including porphyrins, haematoporphyrins, derivatized amino acids, motilins, insulins, angiotensins, peptides, xanthines, purins, purines, nucleosides, nucleotides, oligonucleotides, benzo(a)pyrene-adducted DNAs, catechols, catecholamines, vitamins, barbiturates, antibiotics, steroids, and drugs of abuse. The current optimization of MEKC separations is restricted to empirical approaches, however, because theories for N are incomplete. This study will extend and verify theories for N, by which optimization can be achieved. A family of optimizations with respect to different constraints, e.g., resolution, time, cost, etc., can be envisioned. The research method is based on an analytical model for N that has been verified at low electric field strengths. This theory accounts for contributions to N from longitudinal diffusion and plug size. The theory is fairly sophisticated in its ability to account for variations of mobility and diffusion coefficient with temperature. As the above five parameters are varied, deviations between the model s prediction of N and experimentally determined N s can be related quantitatively to measures of dispersion caused by sources other than longitudinal diffusion and plug size. These measures then can be compared to theoretical estimates of dispersion, as calculated by standard means, to conform or refute the validity of the postulated dispersion mechanism. By implementing a systematic experimental program in conjunction with this model, one should make clear the dependence of N on these parameters.