Modeling Ozone Reaction Rate Constants of Micropollutants Using Quantitative Structure-Property Relationships

A quantitative structure-property relationship model was developed using piecewise linear regression (with two submodels) to predict second-order rate constants with ozone for a wide range of structurally diverse micropollutants. The model was developed using a set of pre-selected structurally diverse micropollutants for which rate constants had been determined experimentally, and then validated with an external data set. A classification function was developed using linear discriminant analysis to classify micropollutants into high-reactive or low-reactive compounds before using the appropriate submodel. The applicability domain of the model was determined using the Williams plot.