Predictive calculation methods for optimization of gradient elution using binary and ternary solvent gradients

Abstract

Predictive optimization methods in gradient elution liquid chromatography are reviewed. The optimization approach is based on calculations of elution volumes and band widths in various modes of gradient elution liquid chromatography. The calculation makes use of the parameters of retention-mobile phase composition equations determined in a few preliminary isocratic experiments. In liquid chromatography with binary solvent gradients, either a direct calculation method may be used to optimise the gradient profile in order to achieve a desired resolution of a “critical pair” of sampls solutes, or “maps” of the dependence of resolution of the individual pairs of compounds on gradient steepness and initial mobile phase composition can be constructed for an assumed constant gradient volume (time). Calculation methods are also applicable to ternary mobile phase gradients in reversed-phase systems. There, ternary “solvent strength”, “selectivity”, or “combined selectivity-solvent strength” gradients are chosen, depending on the separation problem to be solved. In these systems, construction of resolution “maps” is the method of choice of predictive optimization. The precision of predicted gradient elution data depends on the instrumentation used and on its ability to reproduce accurately the gradient profile programme and on possible deviations of the experimental retention-mobile phase composition plots from the theoretically expected forms of these dependences. The effect of the choice of gradient shape (curvature) and volume on the optimized separation is also addressed.