Correlations between gas chromatographic retention data of polycyclic aromatic hydrocarbons and several molecular descriptors

Statistical and factor analysis have been used to test the ability of several molecular descriptors to predict the retention indexes (I) of polycyclic aromatic hydrocarbons determined on SPB-5 and SPB-20 stationary phases. Regression analysis with both columns showed that linear regression with connectivity index and bending energy of the molecules as independent variables gave the best correlation coefficients. ΔI20–5, considered to be a measure of the polar forces in retention, gave a good correlation with the van der Waals radii of the molecules, or strain, and bending energy.