Application of artificial neural networks for the prediction of sulfur polycyclic aromatic compounds retention indices

Quantitative models for structure-retention relationships have been developed for the retention indices of polycyclic aromatic sulfur heterocyclic compounds (PASHs). Six nonlinear models for predicting linear temperature programmed gas chromatographic retention characteristics on a Bpx5 (%5 phenyl) stationary phase. The developed predictive models relate molecular structure of each PASH compound to its experimental retention index.