Predicting the partitioning behavior of various highly fluorinated compounds.

Due to their high degree of fluorination, highly fluorinated compounds (HFCs) have unique substance properties that differ from many other organic contaminants. To predict the environmental behavior of HFCs, models that predict both absorptive and adsorptive partitioning are needed; however, the accuracy of existing models has not heretofore been thoroughly investigated for these compounds. This report has two parts: first we show that a well-established polyparameter linear free energy relationship used to predict experimental adsorption constants underestimates values for HFCs by several orders of magnitude. We found a mechanistic explanation for the model's inaccuracy and adjusted it accordingly. In the second part of this report, we evaluate various models that predict saturated subcooled liquid vapor pressure (pL*), air-water partition constant (Kaw), and the octanol-water partition constant (Kow) based on molecular structure. These parameters are typically required for general environmental fate and transport models. Here, we found that SPARC and COSMOtherm make predictions usually within 1 order of magnitude of the experimental value, while the commonly used EPI SUITE and ClogP perform more inaccurately. The least accurate predictions occurred with ClogP for the fluorotelomer alcohols, where the estimated values were off by 2 to almost 5 orders of magnitude.