Universal Retention Indices and Their Prediction in Reversed-Phase Liquid Chromatography Based On Principal Component Analysis and Target Testing

A universal solute and solvent retention index system for reversed-phase liquid chromatography (LC) has been developed and tested with a library of compounds and mobile phases as the base set. Examination of reversed-phase LC retention data by principal component analysis and target transformation factor analysis reveals that the data obtained from three different reversed-phase columns share a common factor space and that three factors are sufficient to describe these retention data. The resulting eigenvector matrix associated with analyte compounds from singular value decomposition is found to be characteristic of the retention behavior of the compounds and independent of the mobile phases and reversed-phase columns used for the measurement. The same is true for the mobile phase eigenvector matrix. Based on these observations, a reference retention index system is developed for both chromatographic solutes and solvents across different reversed-phase columns. Mean errors of retention prediction using this index system are within 4%.