Inclusion of molecular descriptors in predictive models improves pesticide soil-air partitioning estimates

Environmental factors influencing the soil-air partitioning of semi-volatile petroleum hydrocarbons: Laboratory measurements and optimization model

The soil-air partition coefficient (KSA) values are commonly utilized to examine the fate of organic contaminants in soils; however, their measurement has been lacking for semi-volatile petroleum hydrocarbons within soil contaminated by crude oil. This research utilized a solid-phase fugacity meter to determine the KSA values of n-alkanes and polycyclic aromatic hydrocarbons (PAHs) under crucial environmental conditions. The results showed a notable increase in KSA values with the extent of crude oil contamination in soil. Specifically, in the 3 % crude oil treatment, the KSA values for n-alkanes and PAHs increased by 1.16 and 0.66 times, respectively, compared to the 1 % crude oil treatment. However, the KSA values decreased with changes in temperature, water content, and particle size within the specified experimental range. Among these factors, temperature played a significant role. The KSA values for n-alkanes and PAHs decreased by 0.27-0.89 and 0.61-0.83 times, respectively, with a temperature increase from 5 °C to 35 °C. Moreover, the research identified that the molecular weight of n-alkanes and PAHs contributed to variations in KSA values under identical environmental factors. With an increase in temperature from 5 °C to 35 °C, the range of n-alkanes present in the air phase expanded from C11 to C34, and PAHs showed elevated levels of acenaphthene (ACE) and benzo (b) fluoranthene (BbFA). Furthermore, heightened water content and particle size were observed to facilitate the volatilization of low molecular weight petroleum hydrocarbons. The effect of environmental variables on soil-air partitioning was evaluated using the Box-Behnken design (BBD) model, resulting in the attainment of the lowest log KSA values. These results illustrate that soil-air partitioning is a complex process influenced by various factors. In conclusion, this study improves our comprehension and predictive capabilities concerning the behavior and fate of n-alkanes and PAHs within soil-air systems.

Integrating molecular descriptors for enhanced prediction: Shedding light on the potential of pH to model hydrated electron reaction rates for organic compounds

Hydrated electron reaction rate constant (ke-aq) is an important parameter to determine reductive degradation efficiency and to mitigate the ecological risk of organic compounds (OCs). However, OC species morphology and the concentration of hydrated electrons (e-aq) in water vary with pH, complicating OC fate assessment. This study introduced the environmental variable of pH, to develop models for ke-aq for 701 data points using 3 descriptor types: (i) molecular descriptors (MD), (ii) quantum chemical descriptors (QCD), and (iii) the combination of both (MD + QCD). Models were screened using 2 descriptor screening methods (MLR and RF) and 14 machine learning (ML) algorithms. The introduction of QCDs that characterized the electronic structure of OCs greatly improved the performance of models while ensuring the need for fewer descriptors. The optimal model MLR-XGBoost(MD + QCD), which included pH, achieved the most satisfactory prediction: R2tra = 0.988, Q2boot = 0.861, R2test = 0.875 and Q2test = 0.873. The mechanistic interpretation using the SHAP method further revealed that QCDs, polarizability, volume, and pH had a great influence on the reductive degradation of OCs by e-aq. Overall, the electrochemical parameters (QCDs, pH) related to the solvent and solute are of significance and should be considered in any future ML modeling that assesses the fate of OCs in aquatic environment.

Predicting Concentration- and Ionic-Strength-Dependent Air-Water Interfacial Partitioning Parameters of PFASs Using Quantitative Structure-Property Relationships (QSPRs)

Air-water interfacial retention of poly- and perfluoroalkyl substances (PFASs) is increasingly recognized as an important environmental process. Herein, column transport experiments were used to measure air-water interfacial partitioning values for several perfluoroalkyl ethers and for PFASs derived from aqueous film-forming foam, while batch experiments were used to determine equilibrium K_ia data for compounds exhibiting evidence of rate-limited partitioning. Experimental results suggest a Freundlich isotherm best describes PFAS air-water partitioning at environmentally relevant concentrations (10¹-10⁶ ng/L). A multiparameter regression analysis for K_ia prediction was performed for the 15 PFASs for which equilibrium K_ia values were determined, assessing 246 possible combinations of 8 physicochemical and system properties. Quantitative structure-property relationships (QSPRs) based on three to four parameters provided predictions of high accuracy without model overparameterization. Two QSPRs (R² values of 0.92 and 0.83) were developed using an assumed average Freundlich n value of 0.65 and validated across a range of relevant concentrations for perfluorooctane sulfonate (PFOS), perfluorooctanoate (PFOA), and hexafluoropropylene oxide-dimer acid (i.e., GenX). A mass action model was further modified to account for the changing ionic strength on PFAS air-water interfacial sorption. The final result was two distinct QSPRs for estimating PFAS air-water interfacial partitioning across a range of aqueous concentrations and ionic strengths.

Multiple machine learning algorithms assisted QSPR models for aqueous solubility: Comprehensive assessment with CRITIC-TOPSIS

As an essential environmental property, the aqueous solubility quantifies the hydrophobicity of a compound. It could be further utilized to evaluate the ecological risk and toxicity of organic pollutants. Concerned about the proliferation of organic contaminants in water and the associated technical burden, researchers have developed QSPR models to predict aqueous solubility. However, there are no standard procedures or best practices on how to comprehensively evaluate models. Hence, the CRITIC-TOPSIS comprehensive assessment method was first-ever proposed according to a variety of statistical parameters in the environmental model research field. 39 models based on 13 ML algorithms (belonged to 4 tribes) and 3 descriptor screening methods, were developed to calculate aqueous solubility values (log K_ws) for organic chemicals reliably and verify the effectiveness of the comprehensive assessment method. The evaluations were carried out for exhibiting better predictive accuracy and external competitiveness of the MLR-1, XGB-1, DNN-1, and kNN-1 models in contrast to other prediction models in each tribe. Further, XGB model based on SRM (XGB-1, C = 0.599) was selected as an optimal pathway for prediction of aqueous solubility. We hope that the proposed comprehensive evaluation approach could act as a promising tool for selecting the optimum environmental property prediction methods.

Quantitative assessment of the contribution of soil organic matter functional groups and heteroatoms to PAHs adsorption based on the COSMO-RS model

Soil organic matter (SOM) is considered as a pivotal factor influencing the adsorption of pollutants. However, few prior quantitative investigations of the SOM functional group distribution to the contaminants' fate have been conducted. In this paper, the SOM cluster method based on COSMO-RS theory has been conducted to illustrate the chemical composition variables of SOM that affect the polycyclic aromatic hydrocarbons (PAHs) fate in quantitative terms. In the theoretical simulations, the contributions of carbonyl, carboxyl, aromatic, oxyalkyl and aliphatic groups in SOM to phenanthrene (Phe) and pyrene (Pyr) adsorption are evaluated by calculating the partition coefficients (LogP). The results show that the increase in oxyalkyl content leads to a decrease in LogP. Inversely, carbonyl and carboxyl groups of SOMs positively associated with Phe adsorption. The changes in aromatic and alkyl components have a similar magnitude of influence on LogP. Moreover, the effect of non-carbon-based functional groups in SOM on the Phe partitioning has been examined for the first time. The increase of sulfur and nitrogen content in SOM hinder Phe adsorption, while the rise of phosphorus content promotes the adsorption. In soil adsorption experiments, four natural soils, characterized by X-ray photoelectron spectroscopy (XPS) and Diffuse reflectance infrared Fourier transform (DRIFT), are selected to verify the influence of SOM functional group distribution. Comparing the experimental SOM-water partition coefficient (LogK_oc) with the simulation predicted LogP suggests that the COSMO-RS based SOM cluster method can predict PAHs adsorption ability in SOM.

Versatile in silico modelling of microplastics adsorption capacity in aqueous environment based on molecular descriptor and machine learning

To comprehensively evaluate the hazards of microplastics and their coexisting organic pollutants, the sorption capacity of microplastics is a major issue that is quantified through the microplastic-aqueous sorption coefficient (K_d). Almost all quantitative structure-property relationship (QSPR) models that describe K_d apply only to narrow, relatively homogeneous groups of reactants. Herein, non-hybrid QSPR-based models were developed to predict PE-water (K_PE-w), PE-seawater (K_PE-sw), PVC-water (K_PVC-w) and PP-seawater (K_PP-sw) sorption coefficients at different temperatures, with eight machine learning algorithms. Moreover, novel hybrid intelligent models for predicting K_d more accurately were innovatively developed by applying GA, PSO and AdaBoost algorithms to optimize MLP and ELM models. The results indicated that all three optimization algorithms could improve the robustness and predictability of the standalone MLP and ELM models. In all models trained with K_PE-w, K_PE-sw, K_PVC-w and K_PP-sw data sets, GBDT-1 and XGBoost-1 models, MLP-GA-2 and MLP-PSO-2 models, MLR-3 and MLR-4 models performed better in terms of goodness of fit (R_adj²: 0.907-0.999), robustness (Q_BOOT²: 0.900-0.937) and predictability (R_ext²: 0.889-0.970), respectively. Analyzing the descriptors revealed that temperature, lipophilicity, ionization potential and molecular size were correlated closely with the adsorption capacity of microplastics to organic pollutants. The proposed QSPR models may assist in initial environmental exposure assessments without imposing heavy costs in the early experimental phase.

Percutaneous absorption and exposure risk assessment of organophosphate esters in children's toys

The percutaneous penetration and exposure risk of organophosphate esters (OPEs) from children's toys remains largely unknown. Percutaneous penetration of OPEs was evaluated by EPISkin™ model. Chlorinated OPEs (Cl-OPEs) and alkyl OPEs, except tris(2-ethylhexyl) phosphate, exhibited a fast absorption rate and good dermal penetration ability with cumulative absorptions of 57.6-127 % of dosed OPEs. Cumulative absorptions of OPEs through skin cells were inversely associated with their molecular weight and log octanol-water partition coefficient. Additionally, a quantitative structure-activity relationship model indicated that topological charge and steric features of OPEs were closely related to the transdermal permeability of these chemicals. With the clarification of the factors affecting the transdermal penetration of OPEs, the level and exposure risk of OPEs in actual toys were studied. The summation of 18 OPE concentrations in 199 toy samples collected from China ranged from 6.82 to 228,254 ng/g, of which Cl-OPEs presented the highest concentration. Concentrations of OPEs in toys exhibited clear type differences. Daily exposure to OPEs via dermal, hand-to-mouth contact, and mouthing was evaluated, and dermal contact was a significant route for children's exposure to OPEs. Hazard quotients for noncarcinogenic risk assessment were below 1, indicating that the health risk of OPEs via toys was relatively low.

Effects of temperature and relative humidity on soil-air partition coefficients of organophosphate flame retardants and polybrominated diphenyl ethers

The soil-air partition coefficients (K_SA) of polybrominated diphenyl ethers (PBDEs) and organophosphate flame retardants (OPFRs) is important for determining their fate in soil and air media. However, K_SA values of OPFRs and PBDEs are not available from the current literature, and the effects of environmental factors such as temperature and relative humidity (RH) on K_SA values are not clear. In this study, a solid-phase fugacity meter was used to measure the K_SA values of PBDEs and OPFRs at different temperatures (25, 30, 35, 40, and 45 °C) and relative humidity (RH) conditions (<3 and 100% RH), the relationships between K_SA and octanol-air partition coefficients (K_OA) for OPFRs and PBDEs were analyzed. The results showed that an increase in temperature and RH resulted in a decrease of all K_SA values for PBDEs and OPFRs. Furthermore, the effects of RH on the soil-air partitioning behavior of PBDEs were larger than that of OPFRs. In addition, a significant correlation (p < 0.0001) was observed between log K_SA and log K_OA. The experimental K_SA values of OPFRs and PBDEs were quite different from the predicted K_SA, when calculated with their K_OA values. Overall, this study provides a better understanding for predicting the behavior and fate of OPFRs and PBDEs in soil-air systems.

Prediction models with multiple machine learning algorithms for POPs: The calculation of PDMS-air partition coefficient from molecular descriptor

Polydimethylsiloxane-air partition coefficient (K_PDMS-air) is a key parameter for passive sampling to measure POPs concentrations. In this study, 13 QSPR models were developed to predict K_PDMS-air, with two descriptor selection methods (MLR and RF) and seven algorithms (MLR, LASSO, ANN, SVM, kNN, RF and GBDT). All models were based on a data set of 244 POPs from 13 different categories. The diverse model evaluation parameters calculated from training and test set were used for internal and external verification. Notably, the R_adj², Q_BOOT² and Q_ext² are 0.995, 0.980 and 0.951 respectively for GBDT model, showing remarkable superiority in fitting, robustness and predictability compared with other models. The discovery that molecular size, branches and types of the bonds were the main internal factors affecting the partition process was revealed by mechanism explanation. Different from the existing QSPR models based on single category compounds, the models developed herein considered multiple classes compounds, so that its application domain was more comprehensive. Therefore, the obtained models can fill the data gap of missing experimental K_PDMS-air values for compounds in the application range, and help researchers better understand the distribution behavior of POPs from the perspective of molecular structure.

Prediction of phthalate acid esters degradation in soil using QSAR model: A combined consideration of soil properties and quantum chemical parameters

Phthalic acid esters (PAEs) are predominant hazardous substances and endocrine-disrupting compounds to be controlled in soil. The degradation behaviors of PAEs in soil had been long term concerned. Thus, the degradation rate (K) is important for assessing theexposure risk and is of great significance in evaluating the ecological risk of PAEs in soil environment. But by far, quantitative structure activity relationship (QSAR) models for PAEs degradation have rarely been considered in soil environment. In this study, quantum chemical parameters were considered along with soil properties as two kinds of descriptors in QSAR model. A total of 32 logk of PAEs were collected from reference and experiment. Degradation kinetics in soils were determined by pseudo-first order kinetic models. The residual concentration of PAEs in Udic ferrosols and Aquic cambisols suggesting a potential expose risks of PAEs to ecosystem in soil. The QSAR model between logk and quantum chemical parameters revealed that E_HOMO and qC- are two predominant factors in determining logk value. Furthermore,our study further indicated that soil organic matter (SOM) as new predictor contributes more to predict logk values of PAEs during degradation process than pH. Results from this study make a new contribution for methods to predict the degradation of PAEs in soil environment and highlight the potential to evaluate the environmental risks of degradation of PAEs.

A Review on Prediction Models for Pesticide Use, Transmission, and Its Impacts

The lure of increased productivity and crop yield has caused the imprudent use of pesticides in great quantity that has unfavorably affected environmental health. Pesticides are chemicals intended for avoiding, eliminating, and mitigating any pests that affect the crop. Lack of awareness, improper management, and negligent disposal of pesticide containers have led to the permeation of pesticide residues into the food chain and other environmental pathways, leading to environmental degradation. Sufficient steps must be undertaken at various levels to monitor and ensure judicious use of pesticides. Development of prediction models for optimum use of pesticides, pesticide management, and their impact would be of great help in monitoring and controlling the ill effects of excessive use of pesticides. This paper aims to present an exhaustive review of the prediction models developed and modeling strategies used to optimize the use of pesticides.