Predicting environmental fate parameters with infrared spectroscopy

One of the principal uncertainties associated with risk assessments of organic chemicals in the environment is the lack of chemical-specific values that quantify the many processes determining the chemical's transport and transformation, Because it is not feasible to measure the many constants that are required when thousands of chemicals must be considered, reliable predictive methods are required, A precise theory-based method is now available for many processes whose mechanisms are understood well enough to model theoretically, However, a void exists for processes that are not well understood, We have been developing an infrared spectroscopy-based method to fill that void. Herein, we summarize previously published progress to date, and present new results that strengthen the feasibility of the method, Specifically, we show that the method is capable of predicting constants for ten physical/chemical properties and distribution coefficients for a diverse set of organic esters, typically to within an error percent of ten, The predictive models are built using full infrared spectra and partial least-squares (PLS) modeling, A critical evaluation of the approach, and a brief description of ongoing developments, is presented.