QSRR modelling aimed on the HPLC retention prediction of dimethylamino- and pyrrolidino-substitued esters of alkoxyphenylcarbamic acid

Two homologous series of alkoxyphenylcarbamic acids esters with biological activities were subjected to the QSRR (Quantitative Structure–Retention Relationships) study. Retention factors of studied derivatives were measured in four different HPLC (High Performance Liquid Chromatography) systems. Experimental data in combination with in silico calculated molecular descriptors were used for development complex QSRR model for prediction of retention factor and partially for elucidation of separation mechanisms. The best results in QSRR modelling were given by artificial neural networks in the terms of developing one single robust model with high accuracy of log k prediction ( Q_F3^2 = 0.998) for all studied chromatographic systems. Chemometrical techniques were also used to elucidate the separation mechanisms occurring in the HPLC systems. We found similarities between particular systems and we identified systems which were the most suitable for separation of positional isomers of studied compounds. We observed that solvent significantly influenced the chromatographic separation of positional isomers depending on the used column, whereas combination of acetonitrile and C18-phenyl column caused suppression of π − π interactions, so retention was wholly determined by the hydrophobic interactions.