Predicting electrophoretic mobility of beta-blockers in a water-methanol based electrolyte system

Abstract

Applicability of a mathematical model to predict the electrophoretic mobility of structurally related analytes is shown using generated mobility data for beta-blockers in a mixed aqueous-methanolic buffer The mobilities of atenolol, alprenolol, labetalol and metoprolol in binary mixed solvent buffers containing different concentrations of methanol have been employed to train the model. Then the mobilities of propranolol, timolol and acebutalol have been predicted using the trained model. The produced mean prediction errors employing two and one experimental data points for three beta-blockers are 1.6 and 2.8%, respectively, which is easily within experimental uncertainty.