USING PM3 HAMILTONIAN, FACTOR-ANALYSIS AND REGRESSION-ANALYSIS IN DEVELOPING QUANTITATIVE STRUCTURE-PROPERTY RELATIONSHIPS FOR PHOTOHYDROLYSIS QUANTUM YIELDS OF SUBSTITUTED AROMATIC HALIDES

Based on the descriptors computed by the PM3 hamiltonian and the appli cation of factor analysis and multiple regression analysis, several Qu antitative Structure-Property Relationship (QSPR) equations for disapp earance quantum yields of 45 substituted halogen compounds were obtain ed. After the exclusion of the outliers, the remaining two QSPR equati ons can be used to predict disappearance quantum yields for compounds with similar structure. By stepwise variable selection regression anal ysis, two significant factors relating to the disappearance quantum yi elds of the compounds were identified: the energy of the lowest unoccu pied molecular orbital and the electron-nuclear attraction energy of t he two-center term for carbon-halogen bonds. By factor analysis, other factors that may relate to the quantum yields of the compounds were o btained. The quantum chemical descriptors used in the study can be det ermined by calculation from any given structural formula. So the metho d used in this study permits near a priori prediction of environmental properties for compounds for which the empirical parameters are not r eadily available. (C) 1998 Elsevier Science Ltd. All rights reserved.