Optimization of solvent selectivity for the chromatographic separation of fat-soluble vitamins using a mixture-design statistical technique

A systematic approach, using a mixture-design statistical technique, has been developed for selecting the optimum mobile phase for the separation of fat-soluble vitamins in reversed-phase high-performance liquid chromatography. A quaternary mixture of methanol, acetonitrile, tetrahydrofuran and water was used as mobile phase. Retention time and peak width were recorded in ten runs augmented with five replicates and the data were subsequently fitted to special cubic polynomial models. The resulting mathematical equations enabled prediction of resolution over the entire parameter space. Contour plots of minimum effective resolution and maximum retention time as a function of mobile phase composition are presented and discussed. Visual inspection of these plots provides an overview of the quality of the separation and the analysis time required for each possible mobile-phase composition within the parameter space. It is demonstrated that the methodology followed was an important tool which enabled the taking of informed decisions necessary for selection of the optimum mobile phase for a chromatographic separation. A combination of R-S minimum and t(R) maximum as optimization criteria in a multicriteria decision-making plot using pareto-optimality concept is discussed. This combination enabled visual demonstration of the compromise between separation quality and the economics of analysis time. Our methodology has been compared with the common used technique of overlapping resolution mapping'.