Methods for obtaining accurate predictions of solvent-water partitioning for neutral organic chemicals (e.g., octanol-water partition coefficients) are well established. However, methods that provide comparable accuracy are not available for predicting the solvent-water partitioning of ionic species. Recent work by Franco et al. has demonstrated that for a subset of the 117,000 organic chemicals registered in the European REACH database, approximately 33% have been shown to be “significantly ionized” at environmentally relevant pH values (pH ~ 7.0). Consequently, 1/3 of the chemical database lacks a predictive model for accurately determining the partitioning, and ultimately the fate and transport, of these chemicals in the environment.<br/>This paper outlines a method of predicting solvent-water partition coefficients for ionic species using Abraham pp-LFER solute descriptors estimated from quantum chemistry. For a suite of carboxylic acid anions, solvent-water partition coefficients for 4 solvent-water systems: acetonitrile-, acetone-, methanol-, and dimethylsulfoxide-water (computed from experimental ionization constants in the solvents and water) were predicted with root mean square (RMS) errors of 0.475, 0.512, 0.460, and 0.393, respectively (n = 44, 48, 47, and 41). For a larger set of substituted quaternary amine cations (n = 217), experimentally determined octanol-water partition coefficients were predicted with an RMS error of 1.16.<br/>Predictions made using the quantum-chemically estimated Abraham parameters (QCAPs) were shown to provide improved accuracy in predicting solvent-water partition coefficients, compared to predictions of solvent-water partition coefficients, made using existing Absolv-estimated Abraham solute descriptors derived from the neutral species. For partitioning of anionic solutes in the four organic solvent-water systems, the overall RMS errors were 0.740 and 0.462 for the Absolv and QCAP methods, respectively. For cations partitioning into octanol the  overall RMS errors were 0.997 and 1.16, respectively.<br/>The QCAP method demonstrated improved accuracy over directly-calculated ab initio quantum chemical partition coefficients at comparable levels of theory (M062X/6-31++G**) for both anions partitioning into the 4 organic solvents (RMSE = 0.462 vs. 2.48 for QCAP-predicted vs. direct QC computed, respectively) and cations partitioning into octanol (RMSE =1.16 vs. 2.82 for QCAP-predicted vs. direct QC computed, respectively).<br/>

Predicting solvent-water partitioning of charged organic species using quantum-chemically estimated Abraham pp-LFER solute parameters

254th National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.