Measuring and using scanning-gradient data for use in method optimization for liquid chromatography

The use of scanning gradients can significantly reduce method-development time in reversed-phase liquid chromatography. However, there is no consensus on how they best be used. In the present work we set out to systematically investigate various factors and formulate guidelines. Scanning are used establish retention models for individual analytes. Different were compared by computing Akaike information criterion prediction accuracy. measurement uncertainty was found influence optimum choice model. a third parameter account non-linear relationships consistently not statistically significant. duration (slope) accuracy prediction. error may reduced repeating experiments or – preferably reducing uncertainty. It commonly assumed that gradient-slope factor, i.e. ratio between slopes fastest slowest gradients, should at least three. this factor less important than proximity slope predicted gradient gradients. Also, interpolation preferable extrapolation. For comprehensive two-dimensional chromatography (LC × LC) our results suggest data obtained from fast second-dimension cannot predict much slower first-dimension