Density Functional Theory and Machine Learning-Based Quantitative Structure-Activity Relationship Models Enabling Prediction of Contaminant Degradation Performance with Heterogeneous Peroxymonosulfate Treatments

Machine learning model to predict rate constants for sonochemical degradation of organic pollutants

In this study, machine learning (ML) algorithms were employed to predict the pseudo-1st-order reaction rate constants for the sonochemical degradation of aqueous organic pollutants under various conditions. A total of 618 sets of data, including ultrasonic, solution, and pollutant characteristics, were collected from 89 previous studies. Considering the difference between the electrical power (Pele) and calorimetric power (Pcal), the collected data were divided into two groups: data with Pele and data with Pcal. Eight input variables, including frequency, power density, pH, temperature, initial concentration, solubility, vapor pressure, and octanol-water partition coefficient (Kow), and one target variable of the degradation rate constant, were selected for ML. Statistical analysis was conducted, and outliers were determined separately for the two groups. ML models, including random forest (RF), extreme gradient boosting (XGB), and light gradient boosting machine (LGB), were used to predict the pseudo-1st-order reaction rate constants for the removal of aqueous pollutants. The prediction performance of the ML models was evaluated using different metrics, including the root mean squared error (RMSE), mean absolute error (MAE), and R squared (R2). A significantly higher prediction performance was obtained using data without outliers and augmented data. Consequently, all the applied ML models could be used to predict the sonochemical degradation of aqueous pollutants, and the XGB model showed the highest accuracy in predicting the rate constants. In addition, the power density and frequency were the most influential factors among the eight input variables in prediction with the Shapley additive explanation (SHAP) values method. The degradation rate constants of the two pollutants over a wide frequency range (20-1,000 kHz) were predicted using the trained ML model (XGB) and the prediction results were analyzed.

Machine learning-aided understanding of the structure-activity relationship: a case study of MoS2 supported metal-nonmetal pairs for the hydrogen evolution reaction

Understanding the structure-performance relationship is crucial for designing highly active electrocatalysts, yet this remains a challenge. Using MoS2 supported metal-nonmetal atom pairs (XTM@MoS2, TM = Sc-Ni, and X = B, C, N, O, P, Se, Te, and S) for the hydrogen evolution reaction (HER) as an example, we successfully uncovered the structure-activity relationship with the help of density functional theory (DFT) calculations and integrated machine learning (ML) methods. An ML model based on random forest regression was used to predict the activity, and the trained model exhibited excellent performance with minimal error. SHapley Additive exPlanations analysis revealed that the atom mass and covalent radius of the X atom (m_X and R_X) dominate the activity, and their higher values usually lead to better activity. In addition, four promising candidates, i.e., PCr@MoS2, SV@MoS2, SeTi@MoS2, and SeSc@MoS2, with excellent activity are selected. This work provides several promising catalysts for the HER but, more importantly, offers a workflow to explore the structure-activity relationship.

The molecular binding sequence transformation of soil organic matter and biochar dissolved black carbon antagonizes the transport of 2,4,6-trichlorophenol

Dissolved organic matter (DOM) and dissolved black carbon (DBC) are significant environmental factors that influence the transport of organic pollutants. However, the mechanisms by which their molecular diversity affects pollutant transport remain unclear. This study elucidates the molecular binding sequence and adsorption sites through which DOM/DBC compounds antagonize the transport of 2,4,6-trichlorophenol (TCP) using column experiments and modelling. DBC exhibits a high TCP adsorption rate (kn = 5.32 × 10-22 mol1-n∙Ln-1∙min-1) and conditional stability constant (logK = 5.19-5.74), indicating a strong binding affinity and antagonistic effect on TCP. This is attributed to the high relative content of lipid/protein compounds in DBC (25.65 % and 30.28 %, respectively). Moreover, the small molecule lipid compounds showed stronger TCP adsorption energy (Ead = -0.0071 eV/-0.0093 eV) in DOM/DBC, combined with two-dimensional correlation spectroscopy model found that DOM/DBC antagonized TCP transport in the environment through binding sequences that transformed from lipid/protein small molecule compounds to lignin/tannin compounds. This study used a multifaceted approach to comprehensively assess the impact of DOM/DBC on TCP transport. It reveals that the molecular diversity of DOM/DBC is a critical factor affecting pollutant transport, providing important insights into the environmental trend and ecological effects of pollutants.

Application of machine learning in the study of cobalt-based oxide catalysts for antibiotic degradation: An innovative reverse synthesis strategy

This study addresses antibiotic pollution in global water bodies by integrating machine learning and optimization algorithms to develop a novel reverse synthesis strategy for inorganic catalysts. We meticulously analyzed data from 96 studies, ensuring quality through preprocessing steps. Employing the AdaBoost model, we achieved 90.57% accuracy in classification and an R²value of 0.93 in regression, showcasing strong predictive power. A key innovation is the Sparrow Search Algorithm (SSA), which optimizes catalyst selection and experimental setup tailored to specific antibiotics. Empirical experiments validated SSA's efficacy, with degradation rates of 94% for Levofloxacin and 97% for Norfloxacin, aligning closely with predictions within a 2% margin of error. This research advances theoretical understanding and offers practical applications in material science and environmental engineering, significantly enhancing catalyst design efficiency and accuracy through the fusion of advanced machine learning techniques and optimization algorithms.

Predicting aqueous sorption of organic pollutants on microplastics with machine learning

Microplastics (MPs) are ubiquitously distributed in freshwater systems and they can determine the environmental fate of organic pollutants (OPs) via sorption interaction. However, the diverse physicochemical properties of MPs and the wide range of OP species make a deeper understanding of sorption mechanisms challenging. Traditional isotherm-based sorption models are limited in their universality since they normally only consider the nature and characteristics of either sorbents or sorbates individually. Therefore, only specific equilibrium concentrations or specific sorption isotherms can be used to predict sorption. To systematically evaluate and predict OP sorption under the influence of both MPs and OPs properties, we collected 475 sorption data from peer-reviewed publications and developed a poly-parameter-linear-free-energy-relationship-embedded machine learning method to analyze the collected sorption datasets. Models of different algorithms were compared, and the genetic algorithm and support vector machine hybrid model displayed the best prediction performance (R2 of 0.93 and root-mean-square-error of 0.07). Finally, comparison results of three feature importance analysis tools (forward step wise method, Shapley method, and global sensitivity analysis) showed that chemical properties of MPs, excess molar refraction, and hydrogen-bonding interaction of OPs contribute the most to sorption, reflecting the dominant sorption mechanisms of hydrophobic partitioning, hydrogen bond formation, and π-π interaction, respectively. This study presents a novel sorbate-sorbent-based ML model with a wide applicability to expand our capacity in understanding the complicated process and mechanism of OP sorption on MPs.

Harnessing Data Augmentation and Normalization Preprocessing to Improve the Performance of Chemical Reaction Predictions of Data-Driven Model

As a template-free, data-driven methodology, the molecular transformer model provides an alternative by which to predict the outcome of chemical reactions and design the route of the retrosynthetic plane in the field of organic synthesis and polymer chemistry. However, in consideration of the small datasets of chemical reactions, the data-driven model suffers from the difficulty of low accuracy in the prediction tasks of chemical reactions. In this contribution, we integrate the molecular transformer model with the strategies of data augmentation and normalization preprocessing to accomplish the three tasks of chemical reactions, including the forward predictions of chemical reactions, and single-step retrosynthetic predictions with and without the reaction classes. It is clearly demonstrated that the prediction accuracy of the molecular transformer model can be significantly raised by the use of proposed strategies for the three tasks of chemical reactions. Notably, after the introduction of the 40-level data augmentation and normalization preprocessing, the top-1 accuracy of the forward prediction increases markedly from 71.6% to 84.2% and the top-1 accuracy of the single-step retrosynthetic prediction with additional reaction class increases from 53.2% to 63.4%. Furthermore, it is found that the superior performance of the data-driven model originates from the correction of the grammatical errors of the SMILES strings, especially for the case of the reaction classes with small datasets.