Partition of Neutral Molecules and Ions from Water to o-Nitrophenyl Octyl Ether and of Neutral Molecules from the Gas Phase to o-Nitrophenyl Octyl Ether

Determination of Single-Ion Partition Coefficients between Water and Plasticized PVC Membrane Using Equilibrium-Based Techniques

An experimentally simple method for the direct determination of single-ion partition coefficients between water and a PVC membrane plasticized with o-NPOE is suggested. The method uses the traditional assumption of equal single-ion partition coefficients for some reference cation and anion, in this case tetraphenylphosphonium (TPP+) and tetraphenylborate (TPB-). The method is based on an integrated approach, including direct study of some salts' distribution between water and membrane phases, estimation of ion association constants, and measurements of unbiased selectivity coefficients for ions of interest, including the reference ones. The knowledge of distribution coefficients together with ion association constants allows for direct calculation of the multiple of the single-ion partition coefficients for the corresponding cation and anion, while the knowledge of unbiased selectivity coefficients together with ion association constants allows for immediate estimation of the single-ion partition coefficients for any ion under study, if the corresponding value for the reference ion is known. Both potentiometric and extraction studies are inherently equilibrium-based techniques, while traditionally accepted methods such as voltammetry and diffusion are kinetical. The inner coherent scale of single-ion partition coefficients between water and membrane phases was constructed.

Group Contribution and Machine Learning Approaches to Predict Abraham Solute Parameters, Solvation Free Energy, and Solvation Enthalpy

We present a group contribution method (SoluteGC) and a machine learning model (SoluteML) to predict the Abraham solute parameters, as well as a machine learning model (DirectML) to predict solvation free energy and enthalpy at 298 K. The proposed group contribution method uses atom-centered functional groups with corrections for ring and polycyclic strain while the machine learning models adopt a directed message passing neural network. The solute parameters predicted from SoluteGC and SoluteML are used to calculate solvation energy and enthalpy via linear free energy relationships. Extensive data sets containing 8366 solute parameters, 20,253 solvation free energies, and 6322 solvation enthalpies are compiled in this work to train the models. The three models are each evaluated on the same test sets using both random and substructure-based solute splits for solvation energy and enthalpy predictions. The results show that the DirectML model is superior to the SoluteML and SoluteGC models for both predictions and can provide accuracy comparable to that of advanced quantum chemistry methods. Yet, even though the DirectML model performs better in general, all three models are useful for various purposes. Uncertain predicted values can be identified by comparing the three models, and when the 3 models are combined together, they can provide even more accurate predictions than any one of them individually. Finally, we present our compiled solute parameter, solvation energy, and solvation enthalpy databases (SoluteDB, dGsolvDBx, dHsolvDB) and provide public access to our final prediction models through a simple web-based tool, software packages, and source code.

Development and evaluation of two-parameter linear free energy models for the prediction of human skin permeability coefficient of neutral organic chemicals

The experimental values of skin permeability coefficients, required for dermal exposure assessment, are not readily available for many chemicals. The existing estimation approaches are either less accurate or require many parameters that are not readily available. Furthermore, current estimation methods are not easy to apply to complex environmental mixtures. We present two models to estimate the skin permeability coefficients of neutral organic chemicals. The first model, referred to here as the 2-parameter partitioning model (PPM), exploits a linear free energy relationship (LFER) of skin permeability coefficient with a linear combination of partition coefficients for octanol–water and air–water systems. The second model is based on the retention time information of nonpolar analytes on comprehensive two-dimensional gas chromatography (GC × GC). The PPM successfully explained variability in the skin permeability data (n = 175) with R² = 0.82 and root mean square error (RMSE) = 0.47 log unit. In comparison, the US-EPA’s model DERMWIN™ exhibited an RMSE of 0.78 log unit. The Zhang model—a 5-parameter LFER equation based on experimental Abraham solute descriptors (ASDs)—performed slightly better with an RMSE value of 0.44 log unit. However, the Zhang model is limited by the scarcity of experimental ASDs. The GC × GC model successfully explained the variance in skin permeability data of nonpolar chemicals (n = 79) with R² = 0.90 and RMSE = 0.23 log unit. The PPM can easily be implemented in US-EPA’s Estimation Program Interface Suite (EPI Suite™). The GC × GC model can be applied to the complex mixtures of nonpolar chemicals.