Business-objective-directed, constraint-based multivariate optimization of high-performance liquid chromatography operational parameters

The goal of a separation can be defined in terms of business needs. One goal often used is to provide the required separation in minimum time, but many other goals are also possible. These include maximizing resolution within an analysis-time limit, or minimizing the overall cost. The remaining requirements of the separation can be applied as constraints in the optimization of the goal. We will present a flexible, business-objective-based approach for optimizing the operational parameters of high performance liquid chromatography (HPLC) methods. After selecting the stationary phase and the mobile-phase components, several isocratic experiments are required to build a retention model. Multivariate optimization is performed, within the model, to find the best combination of the parameters being varied so that the result satisfies the goal to the fullest extent possible within the constraints. Interdependencies of parameters can be revealed by plotting the loci of optimal variable values or the function being optimized against a constraint. We demonstrate the concepts with a model separation originally requiring a 54 min analysis time. Multivariate optimization reduces the predicted analysis time to as short as 8 min, depending on the goals and constraints specified.