Evaluation of Some Empirical Retention Models for High Submicellar Liquid Chromatographic Separation of Aromatic Diamines

Abstract

Summary. Description of the retention of five aromatic diamines in high submicellar liquid chromatography was carried out through several hyperbolic and logarithmic retention models using the retention data of mobile phases consisting of sodium dodecyl sulfate (SDS) (0.09-0.15 M) and methanol (50-70% v/v) at pH 3. Among the investigated models, the logarithmic retention modelshowed the best prediction capability and used to predict the solutes retention factors. The mean relative error of the prediction of retention factors of five aromatic diamines in the selected variables space was lower than 0.5%. Based on the best retention model, a grid search program was used to predict the retention times of each solute for all combinations of SDS and methanol concentrations in the factor space. In order to find optimal separation condition, two different chromatographic goals, analysis time and retention differences between adjacent peaks, were evaluated simultaneously using Derringer's desirability function for each mobile phase composition. At the optimal conditions (0.143 M SDS, 53% v/v methanol), a good agreement was observed between predicted and experimental values of the retention times (R 2 = 0.9999 and mean relative error of 0.96%).

Prediction of Analyte Retention Time in Liquid Chromatography

“…Sixteen hyperbolic and logarithmic relationships between log k and both φ and the concentration of sodium dodecyl sulfate monomers ([M]) were examined for the retention of five aromatic diamines by MLC. 184 The best retention model was found to be…”

“…Physicochemical retention modeling has been utilized for micellar liquid chromatography (MLC) , and supercritical fluid chromatography (SFC). Sixteen hyperbolic and logarithmic relationships between log k and both φ and the concentration of sodium dodecyl sulfate monomers ([M]) were examined for the retention of five aromatic diamines by MLC . The best retention model was found to be where c 0 , c 1 , and c 12 are fitting constants.…”