Prediction of Soil Heavy Metal Immobilization by Biochar Using Machine Learning

Machine learning predicting the effects microstructures of biomass hard carbon have on the electrochemical performance of SIBs anodes

This study involved the development of machine learning models to investigate how the microstructures of biomass-based hard carbon (HC) influence sodium storage mechanisms and performance. The goal was to pinpoint the key microstructural features of biomass-based HC that impact the initial coulombic efficiency (ICE) and reversible capacity (RC) of sodium-ion batteries (SIBs). To achieve this, a database was established to correlate structural parameters of HC with essential sodium storage performance metrics (referred to as the Hard Carbon-SIBs, HCSs database). The XGBoost model exhibited high accuracy and excellent generalization in predicting both RC and ICE, achieving coefficients of determination (R2) of 0.88 and 0.77, respectively. SHAP analysis indicated that variations in specific surface area (SSA) significantly affected both electrochemical properties, while PDP analysis identified the key input features influencing RC and ICE. The findings suggest that this methodology holds significant promise for advancing the development of electrochemical energy storage materials.

Investigation of spatially varying relationships between cadmium accumulation and potential controlling factors in the topsoil of island of Ireland based on spatial machine learning approaches

BACKGROUND

Cadmium (Cd) contamination in soils is a pressing environmental issue due to its toxicity and persistence. Given the diverse geological formations and intensive agricultural activities in Ireland, understanding the distribution and sources of soil Cd is particularly important.

METHODS

This study used multiple GIS-based and spatial machine learning (SML) techniques to investigate the spatial distribution and controlling factors of Cd in 16,783 topsoil samples across the island of Ireland. Three analytical methods were applied: hot spot analysis to detect clusters of high and low Cd concentrations, Geographically Weighted Pearson Correlation Coefficients (GWPCC) to explore how Cd relationships with other soil properties vary across space, and Random Forest (RF) to rank the contributing factors in Cd accumulation.

RESULTS

Hot spot analysis revealed strong spatial overlap between Cd concentrations and key geochemical variables including CIA, Fe, P, pH, SOC, and Zn. GWPCC further highlighted their spatially varying relationships, with significantly strong positive correlations between Cd and pH, Zn, and P in the central midlands. The local correlation coefficients obtained from the GWPCC ranged from negative to the highest values of 0.80, 0.92 and 0.86, respectively, which were significantly higher than the results of traditional Pearson correlation coefficients. These patterns were associated with impure limestones, Zn mineralization, and phosphate fertilizer inputs. Furthermore, the RF model ranked Zn (39.4 %) and P (17.6 %) as the most influential factors, with their importance increasing in limestone-dominated areas (50.9 % and 27.4 %), which emphasized the external contributions from local Zn mineralization and phosphate fertilizers in addition to natural accumulation.

CONCLUSION

This study demonstrated the effectiveness of integrating SML techniques with geochemical analysis for identifying Cd sources in the topsoil of Ireland, highlighting the roles of lithology and agricultural activities in Cd accumulation. The results provided valuable insights for contamination management and environmental policy development in Ireland and elsewhere.

Screening estimates of bioaccumulation factors for 4950 per- and polyfluoroalkyl substances in aquatic species

The considerable variability in bioaccumulation factors (BAFs) of per- and polyfluoroalkyl substances (PFAS) across aquatic species, driven by the diversity of PFAS, complex water conditions, and species differences, underscores the resource-intensive nature of relying on experimental data. To develop a robust and effective approach for predicting BAFs, a predictive framework using a three-level stacking deep ensemble learning model was established. Initially, we compiled a substantial dataset of BAFs, encompassing a wide variety of PFAS across both marine and freshwater species. The stacking model demonstrated strong performance, achieving R-squared (R2) values of 0.94 and 0.89, and root-mean-square errors (RMSE) of 0.88 and 1.17 for training and testing, respectively. External validation revealed that 60 % and 90 % of predictions fell within 2-fold and 4-fold differences, respectively, from the observed values. Using this model, we predicted BAFs for 4950 PFAS in 54 global edible fish species, with the predicted median BAF values ranging from 22 L/kg to 477.09 L/kg. The results indicated that PFAS with multiple functional groups (e.g., benzene rings and ketones) exhibited higher BAFs. Finally, an accessible online tool (https://pfasbaf.hhra.net/) was launched to facilitate BAF predictions. This newly released application promises to offer valuable support for environmental risk management and policymaking efforts.

Integration of machine learning and meta-analysis reveals the behaviors and mechanisms of antibiotic adsorption on microplastics

Microplastics (MPs) can adsorb antibiotics (ATs) to cause combined pollution in the environment. Research on this topic has been limited to specific types of MPs and ATs, resulting in inconsistent findings, particularly for the influencing factors and adsorption mechanisms. Therefore, this study combined meta-analysis and machine learning to analyze a dataset comprising 6805 records from 123 references. The results indicated that polyamide has the highest adsorption capacity for ATs, which is primarily attributed to the formation of hydrogen bonds by its N-H groups, and MPs exhibited the strongest affinity for chlortetracycline because the CO and -Cl groups in chlortetracycline form hydrogen and halogen bonds with MPs. Moreover, the particle size, MP and AT concentrations, and pH were key factors affecting the adsorption process with notable interaction effects. Hydrogen bonding and electrostatic interaction were commonly involved in the adsorption of ATs onto MPs. Finally, an interactive graphical user interface was deployed to predict the adsorption amount, affinity constant, and maximum adsorption capacity of MPs for ATs, with results aligning well with the latest published data. This study provides crucial insights into the behavior of MPs carrying ATs, thereby facilitating accurate assessment of the combined environmental risks of them.

Machine learning-guided rare earth recovery from NdFeB magnet waste: Model development, parameter influence analysis and experimental validation

The rapid expansion of the electric vehicle industry has resulted in substantial production of NdFeB magnet wastes from discarded electromotors. These magnets, weighing up to 2 kg in each electromotor, contain 25-35 wt% of strategic rare earth elements (REEs) such as Nd, Pr, and Dy, and their efficient recycling is crucial for sustainable development and environmental protection. Traditional methods for REEs recovery from NdFeB waste, involving oxidizing calcination and acid leaching, require extensive optimization due to waste variability and technological complexities, leading to high costs and environmental risks. Meanwhile, the influence rules of multi-parameters on REEs leaching are complex to comprehensively revealed by the traditional methods. To address these bottlenecks, this study employs machine learning for intelligent REEs recovery from NdFeB waste, bypassing numerous optimization experiments and reveal the complex influencing mechanisms of multi-parameters on REEs leaching. Based on a dataset of 9650 records, the developed model incorporates 24 input features related to waste properties and technological parameters, with 5 outputs corresponding to Nd, Pr, Dy, Co, and Fe leaching efficiencies. Four algorithms were used to develop 20 models to compare their performance. The XGBoost algorithm exhibited the highest prediction accuracy, with R2 values of 0.80-0.99 in the training, test, validation, and 5-fold cross-validation sets. Furthermore, the intricate influencing mechanisms of waste properties, calcination, and acid-leaching parameters on REEs leaching rates was comprehensively elucidated. Finally, a graphical user interface was developed to guide efficient REEs leaching from NdFeB waste and some experiments were conducted to verify its reliability. This study can skip numerous optimization experiments and improve the optimization efficiency, which achieves efficient and intelligent REEs recycling.

Machine learning-assisted prediction of gas production during co-pyrolysis of biomass and waste plastics

A general method for predicting gas yield is crucial in biomass and plastics co-pyrolysis. This study employed two machine learning methods to forecast gas yield in co-pyrolysis. Comparing the predictive performance of Support Vector Regression (SVR) with an R2 of 0.72 and a root mean square error (RMSE) of 0.15, while eXtreme Gradient Boosting (XGBoost) demonstrated a superior performance with an R2 of 0.90 and an RMSE of 0.08. Therefore, XGBoost was selected as the final prediction model. Results obtained from the machine learning interpretation tool, SHapley Additive exPlanations (SHAP), revealed that the two most influential factors affecting gas yield were the highest co-pyrolysis temperature (HTT) and the blending ratio (BR), contributing 33% and 28% to the model's predictions, respectively. Besides, the moisture content in biomass (MB) has been found to be one of the critical variables affecting the gaseous products yield. To determine the interaction between these factors and their contributions to gas yield, SHAP partial dependence analysis (SHAP PDA) was conducted. Therefore, this study offers novel insights into predicting gas yields in biomass and plastics co-pyrolysis, aiding in identifying optimal conditions for maximizing gas yield production.

Interaction and sorption mechanisms of phthalate plasticizers and Cd2+ on biochar

Biochar exhibits significant potential for the remediation of soil contaminated with organic pollutants and heavy metals. A comprehensive understanding of the interfacial interactions and sorption mechanisms of low-hydrophobicity phthalate plasticizers, such as dimethyl phthalate (DMP) and diethyl phthalate (DEP), along with Cd2+ on biochar, is essential for the effective remediation of polluted soil environments. This study systematically examines the interaction and sorption mechanisms of PAEs-Cd2+ on biochar at both macro and micro levels using sorption batch experiments and molecular dynamics simulations. The sorption of contaminants by biochar occurred through a combination of physical and chemical mechanisms. The presence of coexisting pollutants reduced the sorption capacity of biochar to PAEs but had a minimal effect on Cd2+ adsorption. In the co-sorption system, PAEs and Cd2+ demonstrated distinct interaction behaviors. Due to its smaller molecular size and higher diffusion coefficient, Cd2+ readily bonded to surface sorption sites on biochar and infiltrated its pores. Although PAE-ion complexes enhanced the sorption of pollutants by biochar, PAE molecules, and cluster structures primarily accumulated on the biochar surface, interacting with heavy metals through electrostatic forces. This interaction reduced the contribution of pore filling to pollutant sorption and weakened the desorption hysteresis capacity of biochar. The intraparticle diffusion model had similar results. Thus, a larger specific surface area and an abundant pore structure are crucial factors in improving the co-sorption capacity of biochar. This study offers novel insights into the sorption behavior of PAEs and Cd2+ on biochar within organic-inorganic composite pollution.

Manganese oxide application reduces cadmium bioavailability in rice rhizosphere: Insights from desorption kinetics and high-resolution imaging

Cadmium (Cd) contamination in paddy soils threatens global food safety. While manganese (Mn)-based materials show promise in reducing soil Cd bioavailability, their efficacy requires further evaluation. Traditional ex situ sampling methods often overlook metal desorption kinetics and rhizosphere biochemical heterogeneity, potentially misinterpreting Mn's regulatory influence on Cd dynamics. This study employed in situ monitoring tools, including diffusive gradients in thin-films (DGT) measurements, DIFS (DGT-induced fluxes in soils) modeling, and high-resolution DGT and planar optode (PO) imaging, to assess the impact of two Mn oxides (MnO2 and Mn2O3) on Cd bioavailability in rice rhizosphere. Application of MnO2 and Mn2O3 reduced bioavailable Cd by 28.9 % and 15.3 %, respectively, attributed to elevated soil Mn and Fe levels fostering Cd immobilization. DGT-DIFS results revealed that Mn oxide application prolonged Cd replenishment time and reduced its desorption rate from soil solids. PO imaging identified pH heterogeneity in rice rhizosphere, confirming that Mn oxides mediated Cd bioavailability reduction by increasing pH. High-resolution DGT imaging revealed distinct spatial distribution patterns of Cd, Mn, and Fe fluxes, demonstrating Mn's inhibitory effects on Cd bioavailability. These findings highlight the potential of Mn oxides to mitigate Cd uptake by rice, offering a promising strategy for managing Cd-contaminated soils.

Machine Learning-Assisted Molecular Structure Embedding for Accurate Prediction of Emerging Contaminant Removal by Ozonation Oxidation

Ozone has demonstrated high efficacy in depredating emerging contaminants (ECs) during drinking water treatment. However, traditional quantitative structure-activation relationship (QSAR) models often fall short in effectively normalizing and characterizing diverse molecular structures, thereby limiting their predictive accuracy for the removal of various ECs. This study uses embedded molecular structure vectors generated by a graph neural network (GNN), combined with functional group prompts, as inputs to a feedforward neural network. A data set of 28 ECs and 542 data points, representing diverse molecular structures and physiochemical properties, was built to predict the residual rate of ECs (REC) in ozonation oxidation. Compared to traditional QSAR models, the GNN-based molecular structure embedded methods significantly improve prediction accuracy. The resulting KANO-EC model achieved an R2 of 0.97 for REC, demonstrating its ability to capture complex structural features. Moreover, KANO-EC maintains exceptional interpretability, elucidating key functional groups (e.g., carbonyls, hydroxyls, aromatic rings, and amines) involved in the oxidation mechanism. This study presents the KANO-EC model as a novel approach for predicting the ozonation removal efficiency of current and potential ECs. The model also provides valuable insights for developing efficient control strategies for ensuring the long-term safety and sustainability of drinking water supplies.

Sustainable soil remediation using nano-biochar for improved food safety and resource recovery

The contamination of agricultural soils with potentially toxic elements (PTEs) poses serious environmental and health risks due to their persistence and adverse effects on crop productivity. The main objective of this study was to evaluate the potential of nano-biochar (n-BC) to immobilize PTEs in contaminated soil and its effect on PTEs bioaccumulation in lettuce (Lactuca sativa L.), with the hypothesis that n-BC-due to their unique and improved physicochemical properties-are more effective than bulk forms in reducing PTEs mobility and bioavailability. Biochars (BCs) were obtained from palm bunch (PB), rice husk (RH) and sewage sludge (SSL) at 550°C and subsequently processed into nanoscale forms. A six-week pot experiment demonstrated that n-BC amendments significantly reduced the bioavailable (extracted with H2O and CaCl2) fractions of Cr, Cu, Fe, Mn, Ni, Zn, and Pb in soil, with higher immobilization efficiencies by 4.2 % to even 305 % than corresponding bulk biochars (b-BC). According to NICA-Donnan modelling, the main immobilization mechanisms were precipitation and ion exchange. Application of n-BC also resulted in a notable decrease in PTEs concentrations in lettuce leaves (ranging from 29.7 % to 100 %), thereby reducing both the bioaccumulation factor and health risk index. Among the different BCs, SSL-derived n-BC demonstrated the highest immobilization capacity and the most substantial reduction in PTEs uptake by plants. These findings highlight the potential of n-BC as a highly effective and low-cost amendment for rapid mitigation PTEs contamination in agricultural soils, enhancing food safety, and supporting circular economy principles by utilizing organic waste materials.

Co-sorption of phthalate esters and Cd2+ on biochar-sulfhydryl modified montmorillonite composites

Developing efficient passivating sorption materials for the treatment of combined pollution by phthalate esters (PAEs) and Cd2+ has become a critical focus in environmental protection and soil remediation. In this study, biochar-sulfhydryl modified montmorillonite composites (BC-SHMMT) were synthesized, characterized, and evaluated for the co-sorption of PAEs (DMP, DEP) and Cd2+. The integration of sulfhydryl-modified montmorillonite (MMT-SH) with biochar improved the material's surface structure, enhancing both co-sorption capacity and structural stability. Molecular dynamics simulation indicated a monolayer chemical sorption of pollutants by BC-SHMMT, in alignment with the sorption theoretical model. In the co-sorption system, the coexistence of pollutants increased the adsorption of Cd2+ and decreased the sorption of PAEs. The smaller molecular size and higher diffusion coefficient of Cd2+ facilitated its pore filling and interaction with oxygen-containing functional groups of the material, leading to competition with PAEs for sorption sites. Notably, the ion exchange and complexation of Cd2+ adsorption by MMT-SH are pivotal in enhancing BC-SHMMT co-sorption capacity. These findings provide valuable insights into the development of eco-friendly novel materials and the improvement of co-pollution remediation efficiency.

Optimized immobilization of lead and cadmium in soil using dithiocarboxy functionalized silica: A long-term effectiveness study

The efficient immobilization of heavy metals often requires a high dosage of remediation material, resulting in increased remediation costs and potential ecological risks. In this study, we developed a novel silica-based material, RNS-TS, characterized by a high density of dithiocarboxy groups, aimed at remediating Pb and Cd contaminated soils and evaluating the long-term efficacy via aging experiments. The synthesized RNS-TS achieved a functional group density of 2.59 mmol/g. At a concentration of 1 %, it effectively reduced the content of bioaccessibility Pb, Cd, and Cu in slightly contaminated soil by 86 %, 82 %, and 100 %; in moderately contaminated soil by 94 %, 75 %, and 100 %; and in heavily contaminated soil by 68 %, 60 %, and 100 %. Furthermore, the remediation process was relatively fast, with equilibrium achieved within one day after adding the RNS-TS. Aging experiments revealed that the remediated products exhibited excellent stability under simulated climate conditions such as extreme temperatures, freeze-thaw cycles and dry-wet cycles etc. Field experiments demonstrated that a 0.2 wt% application of RNS-TS reduced the content of bioaccessible Cd from 0.6 mg/kg to 0.3 mg/kg (approximately 50.8 % reduction), while Cd content in wheat grains decreased from 0.18 mg/kg to 0.08 mg/kg (approximately 54.4 % reduction). This successful application ensured safe wheat production. This material shows great promise as a risk element stabilization agent for heavy metal-contaminated soils.

Bacterial activation level determines Cd(II) immobilization efficiency by calcium-phosphate minerals in soil

Soil mineral properties significantly influence the mobility of Cd(II) within the soil matrix. However, the limited understanding of how microbial metabolism affects mineral structure at the microscale poses challenges for in situ remediation. Here, we designed a model calcium-phosphate system in a urea-rich environment to explore the impact of different microbial activation levels on Cd(II) fixation at mineral interfaces. Findings indicate that bacteria affected the morphological structure of the minerals and the amount of carbonate incorporation (average 5.4 %), thereby enhancing Cd(II) immobilization capacity (up to 9.6 times). This process is influenced by the intensity of bacterial activation, as reflected in their urease activity. Extracellular substances secreted by bacteria are also essential for activating minerals, contributing to a sustained decrease in their surface potential. The introduction of activated minerals in potting experiments markedly stimulated the soil urease activity, promoting the enrichment of functional bacteria and facilitating Cd(II) passivation, thereby reducing Cd(II) uptake by vegetables. An extensive soil survey further corroborated a linkage between soil total phosphorus and urease activity, indirectly emphasizing the universality of phosphate mineral-urease microbial interactions and their critical role in the morphological transformation of Cd(II). Our findings highlight the functional dynamics of urease microorganisms in shaping soil mineral landscapes and regulating heavy metal mobility, with broad implications for soil microscale remediation strategies.

Insights Into the Efficiency and Health Impacts of Emerging Microplastic Bioremediation Approaches

The pollution caused by microplastics (MPs) is a global environmental and health concern. These plastic particles disrupt food chains and pose health risks to organisms, including humans. From a total of 827 studies, synthetic textiles (35%) and tires (28%) are the primary sources of MPs, with fibers being the most common shape (60%). MPs were detected in feces (44% of studies), lungs (35%), and blood (17%), indicating widespread contamination and potential health impacts. Bioremediation is a promising and sustainable method for mitigating MP pollution, as it uses microorganisms and plants to break down or convert MPs into less hazardous substances. However, it is important to understand and address the potential unintended consequences of bioremediation methods on the environment and human health. This scoping literature review examines the efficiency of currently emerging approaches for microplastic bioremediation, their strengths and weaknesses, and their potential impacts on the environment and human health. Highly effective methods such as mycoremediation, soil microbes for enhanced biodegradation, and phytoextraction were identified, but they pose high toxicity risks. Moderately effective methods include plant-assisted remediation, rhizosphere degradation, phytodegradation, and biodegradation, with effectiveness rates between 50% and 65% and moderate toxicity risks.

Heavy metal adsorption efficiency prediction using biochar properties: a comparative analysis for ensemble machine learning models

The contamination of water and soils with heavy metals poses a significant environmental threat, making the development of effective removal strategies a global priority. Hence, the determination of heavy metals can play an essential role in environmental monitoring and assessment. In the current research, ensemble machine learning (ML) models (i.e., Random Forest Regressor (RFR), Adaptive Boosting (Adaboost), Gradient Boosting (GB), HistGradientBoosting, Extreme Gradient Boosting (XGBoost), and Light Gradient-Boosting Machine (LightGBM)) were applied in attempt to predict the adsorption efficiency of several heavy metals (i.e., Pb, Cd, Ni, Cu, and Zn) according to different factors including temperature, pH, and biochar characteristics. Data were collected from open-source literature review including 353 samples. At the first stage, data processing was performed including outliers' removal and scaling for better data modeling applicability; whereas, in the second stage the predictive models were conducted. The results showed that XGBoost model attained the superior accuracy in comparison with other models by achieving the highest determination coefficient (R2 = 0.92). The research was extended to investigate the feature importance analysis which indicated that the initial concentration ratio of metals to biochar and pH were the most influential factors toward the adsorption efficiency followed by Pyrolysis temperature, while other features like physical properties as surface area and pore structure had a minimal effect on efficiency. These findings highlighted the importance of using ensemble ML models in guiding heavy metals removal solutions as it provides an efficient prediction and ease the selection of the environmental application.

Applications of machine learning in potentially toxic elemental contamination in soils: A review

Soil contamination by potentially toxic elements (PTEs) poses substantial risks to the environment and human health. Traditional investigational methods are often inadequate for large-scale assessments because they are time-consuming, costly, and have a limited accuracy. Machine learning (ML) techniques have emerged as promising tools in environmental studies because of their superiority in processing high-dimensional and unstructured data. However, critical evaluations of contemporary ML applications and methods in PTEs content, distribution, and identification remain scarce. To address this research gap, this study reviews applications of ML to soil PTEs contamination including content prediction, spatial distribution, source identification, and other related tasks. Hyperspectral data combined with ML methods can predict the content of PTEs in large-scale areas at a low cost. In addition, ML algorithms that integrate environmental covariates offer superior performance in spatial predictions compared with traditional geostatistical methods. Moreover, ML techniques incorporated with receptor models provide important advances in the quantitative identification and apportioning of PTE sources, thereby supporting effective environmental management and risk assessment. Based on the frequency of the variables used, we propose that soil pH, soil organic matter (SOM), industrial activities, soil texture, and other relevant factors are key environmental variables that enhance the accuracy of predictions regarding the spatial distribution and source identification of PTEs. From these findings, ML techniques, through their powerful data processing capabilities, provide new perspectives and tools for the efficient assessment and management of soil PTEs contamination.

Stabilization mechanism and remediation effectiveness of Pb and cd in agricultural soil using nonmetallic minerals

Soil contaminated by Pb and Cd has aroused worldwide concern due to the environmental hazards they pose. The effects, mechanisms, and evaluation of Pb and Cd contaminated agricultural soil remediation by nonmetallic minerals are still poorly understood. In this study, solidification/stabilization experiments were used to screen nonmetallic mineral materials and optimize their dosages. Stabilization mechanisms of Pb and Cd by nonmetallic mineral materials were investigated by adsorption kinetics, X-ray diffraction spectroscopy, and Fourier transform infrared spectroscopy. The effectiveness of soil remediation was further confirmed through a pot experiment with pak choi (Brassica rapa L. subsp. chinensis), an important non-heading leafy vegetable. Results demonstrated that the SL composite (composed of 2.5% sepiolite and 1.5% limestone, with a total dosage of 4.0%) exhibits the optimal stabilization effect on soil contaminated with Pb and Cd. In soils with low, medium, and high contamination levels, SL reduced the bioavailability of Pb by 97.97%, 96.78%, and 95.82%, and the bioavailability of Cd by 92.96%, 91.76%, and 91.02%, respectively. SL surfaces are rich in hydroxyl (-OH) and carbonate (CO32-) groups, enabling binding with Pb and Cd ions to form hydroxide and carbonate precipitates. Such interactions suggest that chemical adsorption primarily drives Pb and Cd ion stabilization, reducing their bioavailability in soil. Pak choi grown in SL-remediated soil exhibited Pb and Cd contents compliant with China's food safety standards. These findings further validate the bioavailability reduction rate as a suitable metric for evaluating the remediation effectiveness of heavy metal pollution in agricultural soils. This study provides a new strategy for evaluating the remediation efficiency of heavy metal-contaminated agricultural soil.

Combination of Density Functional Theory and Machine Learning Provides Deeper Insight of the Underlying Mechanism in the Ultraviolet/Persulfate System

The competition between radical and nonradical processes in the activated persulfate system is a captivating and challenging topic in advanced oxidation processes. However, traditional research methods have encountered limitations in this area. This study employed DFT combined with machine learning to establish a quantitative structure-activity relationship between contributions of active species and molecular structures of pollutants in the UV persulfate system. By comparing models using different input data sets, it was observed that the protonation and deprotonation processes of organic molecules play a crucial role. Additionally, the condensed Fukui function, as a local descriptor, is found to be less effective compared to the dual descriptor due to its imprecise definition of f0. The sulfate radical exhibits high selectivity toward local electrophilic sites on molecules, while global descriptors determined by their chemical properties provide better predictions for contribution rates of hydroxyl radicals. Interestingly, there exists a piecewise function relating the contribution rates of different active species to ELU-HO, which is further supported by experimental data. Currently, this relationship cannot be explained by classical chemical theory and requires further investigation. Perhaps this is a new perspective brought to us by combining DFT with machine learning.

Optimisation led energy-efficient arsenite and arsenate adsorption on various materials with machine learning

The contamination of water by arsenic (As) poses a substantial environmental challenge with far-reaching influence on human health. Accurately predicting adsorption capacities of arsenite (As(III)) and arsenate (As(V)) on different materials is crucial for the remediation and reuse of contaminated water. Nonetheless, predicting the optimal As adsorption on various materials while considering process energy consumption continues to pose a persistent challenge. Literature data regarding the As adsorption on diverse materials were collected and employed to train machine learning models (ML), such as CatBoost, XGBoost, and LGBoost. These models were utilized to predict both As(III) and As(V) adsorption on a variety of materials using their reaction parameters, structural properties, and composition. The CatBoost model exhibited superior accuracy, achieving a coefficient of determination (R²) of 0.99 and a root mean square error (RMSE) of 1.24 for As(III), and an R² of 0.99 and RMSE of 5.50 for As(V). The initial As(III) and As(V) concentrations were proved to be the primary factors influencing adsorption, accounting for 27.9 % and 26.6 % of the variance for As(III) and As(V) individually. The genetic optimization led optimisation process, considering the low energy consumption, determined maximum adsorption capacities of 291.66 mg/g for As(III) and 271.56 mg/g for As(V), using C-Layered Double Hydroxide with reduced graphene oxide and chitosan combined with rice straw biochar, respectively. To further facilitate the process design for different real-life applications, the trained ML models are embedded into a web-app that the user can use to estimate the As(III) and As(V) adsorption under different design conditions. The utilization of ML for the energy-efficient As(III) and As(V) adsorption is deemed essential for advancing the treatment of inorganic As in aquatic settings. This approach facilitates the identification of optimal adsorption conditions for As in various material-amended waters, while also enabling the timely detection of As-contaminated water.

Machine learning-based prediction of non-aeration linear alkylbenzene sulfonate mineralization in an oxygenic microalgal-bacteria biofilm

Microalgal-bacteria biofilm shows great potential in low-cost greywater treatment. Accurately predicting treated greywater quality is of great significance for water reuse. In this work, machine learning models were developed for simulating and predicting linear alkylbenzene sulfonate (LAS) removal using 152-days collected data from a battled oxygenic microalgal-bacteria biofilm reactor (MBBfR). By using nine variables including influent LAS, hydraulic retention time (HRT), biofilm density and thickness, specific oxygen production and consumption rates, microalgae and bacteria concentrations, and dissolved oxygen (DO), the support vector machine (SVM) model enabled the accurate LAS removal prediction (training set: R2 = 0.995, (root mean square error, RMSE) = 0.076, (mean absolute error, MAE) = 0.069; testing set: R2 = 0.961, RMSE = 0.251, MAE = 0.153). SVM can be also successfully applied for MBBfR operation optimization (HRT = 4.28 h, DO = 0.25 mg/L) that achieving accurate prediction of LAS mineralization.

Potassium permanganate-hematite-modified biochar enhances cadmium and zinc passivation and nutrient availability and promotes soil microbial activity in heavy metal-contaminated soil

Modified biochar has garnered considerable attention for its versatile applications in remediating soils contaminated with heavy metals. However, most existing studies have primarily focused on the stabilisation of heavy metals, with limited research exploring the broader environmental effects following the application of modified biochar. In this study, we developed a potassium permanganate (KMnO4)-hematite-modified biochar (MnFeB) as a passivating agent for heavy metals, specifically targeting cadmium (Cd) and zinc (Zn)-contaminated soils. We examined the effects of MnFeB on the biotoxicity of Cd and Zn, soil properties, enzyme activities, heavy metal resistance genes (czcA, czcC, and czcD), and the soil microbial community in contaminated soils. Treatment with MnFeB markedly reduced the soil diethylenetriaminepentaacetic acid (DTPA)-extractable Zn and Cd contents by 18.79% and 43.65%, respectively. Furthermore, soil organic carbon (SOC), cation exchange capacity (CEC), and the availability of nitrogen, phosphorus, and potassium were found to be increased. MnFeB application also enhanced the activities of catalase, urease, and alkaline phosphatase while reducing the expression of czcA by 23.63%. Moreover, changes in the composition and diversity of soil bacterial and fungal communities were observed. These findings highlight the effects of environmental changes induced by MnFeB application on Cd/Zn-contaminated soil and offer theoretical support for employing passivation strategies in the remediation of heavy metal-contaminated soils.

Machine learning-based analysis of microplastic-induced changes in anaerobic digestion parameters influencing methane yield

Microplastics (MPs) present significant challenges for anaerobic digestion (AD) processes used in energy recovery from contaminated organic waste. Given that optimal AD conditions vary widely across studies when MPs are present, a robust predictive model is essential to accurately assess these complex effects. This study applied four machine learning algorithms to predict methane yield using two datasets-one with and one without MPs. Among these, gradient boosting regression demonstrated the highest prediction accuracy, with testing R2 values of 0.996 for systems without MP pollution and 0.998 with MP pollution. This model was then further optimized by removing redundant and low-importance features, refining its predictive power. Feature importance analysis revealed that digestion time and substrate organic matter content were key parameters positively correlated with methane production. In the presence of MPs, substrate pH and inoculum total solids emerged as critical factors, with partial dependence plots offering deeper insights into their optimal conditions. This research offers new perspectives on the intricate effects of MPs on methane production, which could inform the optimization of AD processes in environments contaminated by MPs.

Intelligent leaching of Zn and Mn from spent disposable batteries to avoid traditional optimizing experiments

Spent disposable Zn-Mn and Zn-C batteries are important resources for recycling. Acid leaching is the crucial step in the hydrometallurgy process for recycling Zn and Mn from these spent Zn-based batteries. However, to obtain the optimal leaching efficiency, the uncontrollable components in waste feed and various leaching parameters cause numerous replicated optimal experiments, increasing the recovery cost and environmental risks. To solve the issues, we employed machine learning (ML) techniques to construct models to predict Zn and Mn leaching from spent disposable batteries without optimizing experiments. Among four ML algorithms tested, the extreme gradient boosting demonstrated superior predictive performance, achieving an R2 of 0.85-0.98 across the training, test, and verification datasets. An analysis of feature importance indicated that the particle size, waste composition, acid concentration, temperature, and time affected the metal leaching most. This study also revealed the interaction effects of the waste properties and leaching process on the metal leaching. Furthermore, we created a user-friendly graphical user interface (GUI) that enables quick acquisition of metal leaching results, requiring only the measurement of waste particle size and component. Finally, experimental verification confirmed the practicability of the GUI. This study achieves intelligent metal leaching from spent batteries and overcomes the high recovery cost and environmental risks associated with traditional experimental optimizing methods.

Heavy metals release in lead-zinc tailings: Effects of weathering and acid rain

Heavy metals (HMs) release from lead (Pb)-zinc (Zn) tailings poses significant environmental risks to surrounding areas. Furthermore, with the natural weathering and frequently happened acid rain events, the release of HMs could be elevated. This study conducted a series of laboratory column experiments with thermodynamics and hydrogeochemical analysis to investigate the environmental behavior of HMs release in Pb-Zn tailings under natural weathering conditions and acid rain events. Results showed that the weathering of calcite facilitates the release of Pb (17.9 mg/kg) and cadmium (Cd) (0.15 mg/kg), while acid rain promotes Zn release (10.5 mg/kg) from the Fe-Mn oxides, with no significant change for arsenic (As). Among the influencing factors during the column experiments, the oxidation-reduction potential (ORP) was identified as the primary indicator for the predictions of the HMs release behavior based upon the Random Forest model (R2 = 0.973 - 0.997). Correlation analysis revealed a strong relationship between coexistent ions and HM release patterns. Therefore, saturation index (SI) could effectively identify the influence range of each mineral phase on HM release. This study provides scientific evidence for effective management in carbonate-type tailings ponds.

Probabilistic prediction of phosphate ion adsorption onto biochar materials using a large dataset and online deployment

Phosphate (PO4(III)) contamination in water bodies poses significant environmental challenges, necessitating efficient and accurate methods to predict and optimize its removal. The current study addresses this issue by predicting the adsorption capacity of PO4(III) ions onto biochar-based materials using five probabilistic machine learning models: eXtreme Gradient Boosting LSS (XGBoostLSS), Natural Gradient Boosting, Bayesian Neural Networks (NN), Probabilistic NN, and Monte-Carlo Dropout NN. Utilizing a dataset of 2952 data points with 16 inputs, XGBoostLSS demonstrated the highest R2 (0.95) on new adsorbents. SHapely Additive exPlanations analysis showed that adsorption experimental conditions had the most significant impact (43%), followed by synthesis conditions (29%) and adsorbent characteristics (28%). Optimized conditions included an initial PO4(III) concentration of 125 mg/L, carbon content of 11.5%, oxygen content of 23%, a contact time of 1440 min, a heating rate of 5 °C/min, 200 rpm, and a surface area of 410 m2/g, using Ra-LDO adsorbent synthesized from rape cabbage feedstock. This study developed and presented a practical online framework for predicting PO4(III) removal onto biochar using a web-based graphical user interface.

Natural factor-based spatial prediction and source apportionment of typical heavy metals in Chinese surface soil: Application of machine learning models

Predicting the natural distribution of heavy metals (HMs) in soil is important to understand the potential risk of pollution. However, suitable technologies are still lacking for wide scale due to the large spatial heterogeneity. In this study, we developed machine learning models for predicting natural contents of five typical HMs in soil, including As, Cd, Cr, Hg and Pb. It was found that the optional random forest (RF) model had the best performance with the R2 up to 0.64. Based on this model, potential distribution of the five HMs explored that elevated contents were mainly concentrated in the southwest and south central of China. Feature analysis illustrated that importance of natural factors followed the order of geological attributes > soil properties > climatic conditions > ecological functions. In particular, lithology of the parent material dominated the content of metals, with the contributions of 18-25%. Moreover, soil properties of pH, cation exchange capacity, profile depth of soil and vegetation coverage had different influences on HMs, due to the variability in the properties of different HMs. This study developed a mapping relationship between natural factors and soil HMs by data science method, which may provide instructive information for pollution control and planning decisions.

Design optimization of bimetal-modified biochar for enhanced phosphate removal performance in livestock wastewater using machine learning

Mg-modified biochar shows high adsorption performance under weakly acidic and neutral water conditions. However, its phosphate removal efficiency markedly decreases in naturally alkaline wastewater, such as that released in livestock farming (anaerobic wastewater with a high phosphate concentration). This research employed six machine learning models to predict and optimize the phosphate removal performance of bimetal-modified biochar (i.e., Mg-Ca/Al/Fe/La) to develop material design strategies suitable for achieving high removal efficiency in alkaline wastewater. Random forest, gradient boosting regressor, and extreme gradient boosting models achieved high prediction accuracy (R2 > 0.98). Model predictions and experimental validations indicated that Mg-Ca-modified biochar still maintained high adsorption capacity under acidic conditions and could effectively realize phosphate adsorption under alkaline conditions, with a removal rate of 99.33 %. Overall, this research focuses on material performance optimization using machine learning, offering insights and methods for developing biochar materials for practical water-treatment applications.

Metal contamination - a global environmental issue: sources, implications & advances in mitigation

Metal contamination (MC) is a growing environmental issue, with metals altering biotic and metabolic pathways and entering the human body through contaminated food, water and inhalation. With continued population growth and industrialisation, MC poses an exacerbating risk to human health and ecosystems. Metal contamination in the environment is expected to continue to increase, requiring effective remediation approaches and harmonised monitoring programmes to significantly reduce the impact on health and the environment. Bio-based methods, such as enhanced phytoextraction and chemical stabilisation, are being used worldwide to remediate contaminated sites. A systematic plant screening of potential metallophytes can identify the most effective candidates for phytoremediation. However, the detection and prediction of MC is complex, non-linear and chaotic, and it frequently overlaps with various other constraints. Rapidly evolving artificial intelligence (AI) algorithms offer promising tools for the detection, growth and activity modelling and management of metallophytes, helping to fill knowledge gaps related to complex metal-environment interactions in different scenarios. By integrating AI with advanced sensor technologies and field-based trials, future research could revolutionize remediation strategies. This interdisciplinary approach holds immense potential in mitigating the detrimental impacts of metal contamination efficiently and sustainably.

Machine learning helps reveal key factors affecting tire wear particulate matter emissions

Tire wear particles (TWPs) are generated with every rotation of the tire. However, obtaining TWPs under real driving conditions and revealing key factors affecting TWPs are challenging. In this study, we obtained a TWPs dataset by simulating tire wear process under real driving conditions using a tire wear simulator and custom-designed test conditions. This study shows that tire wear PM2.5 accounts for about 65 % of PM10. The response relationship between TWP emissions (both PM2.5 and PM2.5-10) and factors (the radial force, the lateral force, the tangential force, speed, driving torque, tire contact area, total contour length and tire tread temperature) was obtained by machine learning (ML) method. The random forest (RF) model was developed and displayed good prediction performance with an R2 of 0.84 and 0.78 for PM2.5 and PM2.5-10 on the test set, respectively. Model-related (similarity network graph) and model-unrelated (partial dependence plots and centered-individual conditional expectation plots) explainability methods were used to break the black box of ML. Model explainability results show that the feature parameters-emission response relationships for tire wear PM2.5 and PM2.5-10 are different. Avoiding strenuous driving behaviors (TTF < 400 N, TLF < 400 N), reducing tread temperature (T < 45℃), and minimizing the number of small tread patterns are feasible ways to reduce TWPs.

Predicting the efficiency of arsenic immobilization in soils by biochar using machine learning

Arsenic (As) pollution in soils is a pervasive environmental issue. Biochar immobilization offers a promising solution for addressing soil As contamination. The efficiency of biochar in immobilizing As in soils primarily hinges on the characteristics of both the soil and the biochar. However, the influence of a specific property on As immobilization varies among different studies, and the development and application of arsenic passivation materials based on biochar often rely on empirical knowledge. To enhance immobilization efficiency and reduce labor and time costs, a machine learning (ML) model was employed to predict As immobilization efficiency before biochar application. In this study, we collected a dataset comprising 182 data points on As immobilization efficiency from 17 publications to construct three ML models. The results demonstrated that the random forest (RF) model outperformed gradient boost regression tree and support vector regression models in predictive performance. Relative importance analysis and partial dependence plots based on the RF model were conducted to identify the most crucial factors influencing As immobilization. These findings highlighted the significant roles of biochar application time and biochar pH in As immobilization efficiency in soils. Furthermore, the study revealed that Fe-modified biochar exhibited a substantial improvement in As immobilization. These insights can facilitate targeted biochar property design and optimization of biochar application conditions to enhance As immobilization efficiency.

A Review on Remediation Technology and the Remediation Evaluation of Heavy Metal-Contaminated Soils

With the rapid development of industry and agriculture, soil contamination has become a significant environmental issue, and the heavy metal contamination of soils is an important part of it. The main methods for the remediation of heavy metal-contaminated soils include physical methods, chemical methods, biological methods, and combined remediation methods have been proposed as research deepens. However, the standards and evaluation methods for the remediation of heavy metal-contaminated soils are still not well-established. This article discusses the sources and contamination status of heavy metals in soils, the advantages and disadvantages of remediation technology for heavy metal-contaminated soils, remediation standards, and post-remediation evaluation methods. It also proposes scientific issues to be addressed in future research and provides an outlook on future development, hoping to assist in subsequent remediation studies of heavy metal-contaminated soils.

Predicting bioavailable barium transfer in soil-bok choy systems: A study induced by shale gas extraction in Chongqing, China

Barium (Ba) is a significant contaminant from shale gas extraction and is also used in various other industries. However, there has been very limited attention paid to Ba. Elucidating the Ba in soil-crop system are of great significance for both human health risk assessment and pollution control. In this study, the bioavailability of Ba in soils was studied by using various characterization methods. Then the major factors dominating the transfer of Ba in soil-bok choy system and a suitable predicted model was derived. The results showed that Ba was mainly accumulated in the roots (transfer factor < 0.3). The relationships between Ba in shoots and the bioavailability of Ba characterizing with different methods increased in the order of CH3COOH (R2 = 0.81) < ethylenediamine tetraacetic acid (R2 = 0.87) < pore water (R2 = 0.89) < diffusive gradients in thin film (R2 = 0.90) < CaCl2 (R2 = 0.91). The major soil properties affecting Ba in shoots were pH (r = -0.32, P > 0.05), cation exchange capacity (r = -0.43, P < 0.01) and labile Al (r = 0.38, P < 0.05). Bioavailability of Ba can preferably model the Ba transfer in soil-bok choy system. The best reliable model was LogBa[shoot] = 0.591LogBa[soil-Pore water] + 1.749 (R2 = 0.963, P < 0.001). This model without measuring soil physicochemical properties, making it easier and more convenient to use in practice. Overall, these results highlight the role of metal bioavailability in predicting their transfer in soil-plant systems.

A novel hybrid variable cross layer-based machine learning model improves the accuracy and interpretation of energy intensity prediction of wastewater treatment plant

Energy intensity (EI) prediction in wastewater treatment plants (WWTPs) suffers from inaccuracy and non-interpretability due to poor data quality, complex mechanisms and various confounding variables. In this study, the novel hybrid variable cross layer-based machine learning (VCL-ML) model was devised, which generates new knowledge with monitoring indicators (e.g., COD, etc.) and then embeds both domain knowledge and monitoring indicators into the ML model. This novel hybrid VCL-ML model achieves a root-mean-square error (RMSE) of 0.021 kW h/m³ with an 8.7% improvement over the conventional ML (Con-ML) model. The Shapley additive explanation demonstrated that domain knowledge features are ranked high and have important interpretable implications for the model, such as capacity utilization (CU), which measures the efficiency of resource use, and total nitrogen remaining rate (TN_rr), which indicates the nitrogen retention in a system. Partially dependent interactions between domain knowledge (e.g., sludge yield) and monitoring indexes (e.g., influent pH) could contribute to the interpretation of reality. By comparing the feature categorization between VCL-ML and Con-ML models, temporal information (e.g., month) and removal information (e.g., TN_rr) played an important role in the model's performance improvement. This result highlights the strong correlation between wastewater treatment plant energy intensity with pollutant removal and temporal information while weakening the contribution of other redundant features. This VCL-ML model improves the predicting accuracy and interpretation of the EI of WWTPs, which can be used in the optimal operation and sustainable management of WWTPs.

Machine learning prediction of fundamental sewage sludge biochar properties based on sludge characteristics and pyrolysis conditions

Sewage sludge biochar (SSBC) has significant potential for resource recovery from sewage sludge (SS) and has been widely studied and applied across various fields. However, the variability in SSBC properties, resulting from the diverse nature of SS and its intricate interaction with pyrolysis conditions, presents notable challenges to its practical use. This research employed machine learning techniques to predict fundamental SSBC properties, including elemental content, proximate compositions, surface area, and yield, which are essential for assessing the applicability of SSBC. The models achieved robust predictive accuracy (test R2 = 0.82-0.95), except for surface area. Notably, analysis of the optimal models revealed SS characteristics had a significant impact on SSBC properties, particularly total and fixed carbon content (combined importance exceeding 80%). This underscores the needs of source analysis and preparation optimization in targeted SS recovery or SSBC applications. To facilitate this, a graphical user interface was developed for strategic analyzation of sludge sources and pyrolysis settings.

Advances and future directions of environmental risk research: A bibliometric review

Environmental risk is one of the world's most significant threats, projected to be the leading risk over the next decade. It has garnered global attention due to increasingly severe environmental issues, such as climate change and ecosystem degradation. Research and technology on environmental risks are gradually developing, and the scope of environmental risk study is also expanding. Here, we developed a tailored bibliometric method, incorporating co-occurrence network analysis, cluster analysis, trend factor analysis, patent primary path analysis, and patent map methods, to explore the status, hotspots, and trends of environment risk research over the past three decades. According to the bibliometric results, the publications and patents related to environmental risk have reached explosive growth since 2018. The primary topics in environmental risk research mainly involve (a) ecotoxicology risk of emerging contaminants (ECs), (b) environmental risk induced by climate change, (c) air pollution and health risk assessment, (d) soil contamination and risk prevention, and (e) environmental risk of heavy metal. Recently, the hotspots of this field have shifted into artificial intelligence (AI) based techniques and environmental risk of climate change and ECs. More research is needed to assess ecological and health risk of ECs, to formulize mitigation and adaptation strategies for climate change risks, and to develop AI-based environmental risk assessment and control technology. This study provides the first comprehensive overview of recent advances in environmental risk research, suggesting future research directions based on current understanding and limitations.

Machine learning-based identification of critical factors for cadmium accumulation in rice grains

The aggregation of Cadmium (Cd) in rice grains is a significant threat to human healthy. The complexity of the soil-rice system, with its numerous influencing parameters, highlights the need to identify the crucial factors responsible for Cd aggregation. This study uses machine learning (ML) modeling to predict Cd aggregation in rice grains and identify the influencing factors. Data from 474 data points from 77 published works were analyzed, and eight ML models were established using different algorithms. The input variables were total soil Cd concentration (TS Cd) and extractable Cd concentration (Ex-Cd), while rice Cd concentration (Cdrice) was the output variable. Among the models, the Extremely Randomized Trees (ERT) model performed the best (TS Cd: R2 = 0.825; Ex-Cd: R2 = 0.792), followed by Random Forest (TS Cd: R2 = 0.721; Ex-Cd: R2 = 0.719). The ERT feature importance ranking analysis revealed that the essential factors responsible for Cd aggregation are cation exchange capacity (CEC), TS Cd, Water Management Model (WMM), and pH for total soil Cd as input variables. For extractable Cd as an input variable, the vital factors are CEC, Ex-Cd, pH, and WMM. The study highlights the importance of the Water Management Model and its impact on Cd concentration in rice grains, which has been overlooked in previous research.Please check and confirm that the authors and their respective affiliations have been correctly identified and amend if necessary.The authors and their respective affiliations are correct.Author details: Kindly check and confirm whether the corresponding author is correctly identified.It is correct.

Identifying the pollution risk pattern from industrial production to rural settlements and its countermeasures in China

Impacted by large-scale and rapid rural industrialization in the past few decades, China's rural settlements are confronted with the risk of heavy metal pollution stemming from industrial production, which might pose a significant threat to the rural habitat and the well-beings. This study devised a relative risk model for industrial heavy metal pollution to the rural settlements based on the source-pathway-receptor risk theory. Using this model, we assessed the risk magnitudes of heavy metal pollution from industrial production at a 10 km × 10 km grid scale and identified the characteristics of the risk pattern in China. Our finding reveals: (1) the relative risk values of wastewater, waste gas and total heavy metal pollution are notably concentrated within a confined spectrum, with only a small number of units are characterized by high-risk level; (2) Approximately 21.57 % of China's rural settlements contend with heavy metal pollution, with 4.17 %, 9.84 % and 7.55 % being subjected to high, medium and low risks, respectively; (3) The high-risk units mainly is concentrated in the developed areas such as Yangtze River Delta, Pearl River Delta, and the Beijing-Tianjin metropolitan area, also dispersed in the plain areas with high rural population density. Guided by these insights, this study puts forth regionally tailored prevention and control strategies, as well as distinct process prevention and control strategies.

Mining soil heavy metal inversion based on Levy Flight Cauchy Gaussian perturbation sparrow search algorithm support vector regression (LSSA-SVR)

Soil heavy metal pollution in mining areas poses severe challenges to the ecological environment. In recent years, machine learning has been widely used in heavy metal inversion by hyperspectral data. However, deterministic algorithms and probabilistic algorithms may confront local optimal solutions in practical applications. The local optimal solution is not the optimal value obtained within the entire defined interval, and as a result will affect the reliability of these approaches. This paper proposes a Levy Flight Cauchy Gaussian perturbation Sparrow Search algorithm Support Vector Regression (LSSA-SVR) soil heavy metal content prediction model. It introduces Levy Flight (LF) measurement and Cauchy Gaussian perturbation based on the Sparrow search algorithm. The LSSA-SVR model was shown to increase the breadth of solutions searched, avoiding the local optimal solution problem. When applied to mining soil heavy metal experiments, we found that the LSSA-SVR model gave a good fit for the elements Cu, Zn, As, and Pb. The correlation coefficients between the predicted results and the actual results of the four elements were all above 0.94. The heavy metal predicted results of LSSA-SVR have a small error margin in both the overall distribution and in individual differences. This study provides an efficient and accurate monitoring method for mining soil heavy metal inversion. It also provides strong support for environmental management and soil remediation.

Adsorption and migration behaviors of heavy metals (As, Cd, and Cr) in single and binary systems in typical Chinese soils

In this study, the competitive adsorption and migration behaviors of arsenic (As), cadmium (Cd), and chromium (Cr) in typical Chinese soils were investigated. It was observed that Hainan, Shanxi, and Zhejiang Mengjiadai soils exhibited the highest adsorption capacities for As (563 μg/g), Cd (653 μg/g), and Cr (383 μg/g), respectively. Heavy metals (HMs) adsorption capacities were predicted by Extreme gradient boosting (XGBoost) models, and the Shapley additive explanation (SHAP) was employed to elucidate the effect of soil physicochemical properties on target values. Due to redox and complexation reaction, the primary factor affecting adsorption has changed from free state manganese (Mn) in single As system to antimony (Sb) in As/Cd and As/Cr systems. Furthermore, the maximum adsorption capacity (Qm) of As increased by 49.4 % with the addition of Cd into Heilongjiang soil. Finally, the migration process of HMs in Heilongjiang, Hebei, and Hainan soils was simulated by column experiments. With a relatively large dispersion coefficient (D = 29.630 cm2/h) and small retardation factor (Rh = 0.030), Cr penetrated fastest in Heilongjiang soil. This research demonstrates that both the types and coexistence of HMs may affect the HMs behaviors in soil.

Novel nitrogen-rich hydrogel adsorbent for selective extraction of rare earth elements from wastewater

Efficient recovery of rare earth elements (REEs) from wastewater is crucial for advancing resource utilization and environmental protection. Herein, a novel nitrogen-rich hydrogel adsorbent (PEI-ALG@KLN) was synthesized by modifying coated kaolinite-alginate composite hydrogels with polyethylenimine through polyelectrolyte interactions and Schiff's base reaction. Various characterizations revealed that the high selective adsorption capacity of Ho (155 mg/g) and Nd (125 mg/g) on PEI-ALG@KLN is due to a combination of REEs (Lewis acids) via coordination interactions with nitrogen-containing functional groups (Lewis bases) and electrostatic interactions; its adsorption capacity remains more than 85 % after five adsorption-desorption cycles. In waste NdFeB magnet hydrometallurgical wastewater, the recovery rate of PEI-ALG@KLN for Nd and Dy can reach more than 93 %, whereas that of Fe is only 5.04 %. Machine learning prediction was used to evaluate adsorbent properties via different predictive models, with the random forest (RF) model showing superior predictive accuracy. The order of significance for adsorption capacity was pH > time > initial concentration > electronegativity > ion radius, as indicated by the RF model feature importance analysis and SHapley Additive exPlanations values. These results confirm that PEI-ALG@KLN has considerable potential in the selective extraction of REEs from wastewater.

Refining hydrogel-based sorbent design for efficient toxic metal removal using machine learning-Bayesian optimization

Hydrogel-based sorbents show promise in the removal of toxic metals from water. However, optimizing their performance through conventional trial-and-error methods is both costly and challenging due to the inherent high-dimensional parameter space associated with complex condition combinations. In this study, machine learning (ML) was employed to uncover the relationship between the fabrication condition of hydrogel sorbent and their efficiency in removing toxic metals. The developed XGBoost models demonstrated exceptional accuracy in predicting hydrogel adsorption coefficients (Kd) based on synthesis materials and fabrication conditions. Key factors such as reaction temperature (50-70 °C), time (5-72 h), initiator ((NH4)2S2O8: 2.3-10.3 mol%), and crosslinker (Methylene-Bis-Acrylamide: 1.5-4.3 mol%) significantly influenced Kd. Subsequently, ten hydrogels were fabricated utilizing these optimized feature combinations based on Bayesian optimization, exhibiting superior toxic metal adsorption capabilities that surpassed existing limits (logKd (Cu): increased from 2.70 to 3.06; logKd (Pb): increased from 2.76 to 3.37). Within these determined combinations, the error range (0.025-0.172) between model predictions and experimental validations for logKd (Pb) indicated negligible disparity. Our research outcomes not only offer valuable insights but also provide practical guidance, highlighting the potential for custom-tailored hydrogel designs to combat specific contaminants, courtesy of ML-based Bayesian optimization.

Predicting removal of arsenic from groundwater by iron based filters using deep neural network models

Arsenic (As) contamination in drinking water has been highlighted for its environmental significance and potential health implications. Iron-based filters are cost-effective and sustainable solutions for As removal from contaminated water. Applying Machine Learning (ML) models to investigate and optimize As removal using iron-based filters is limited. The present study developed Deep Learning Neural Network (DLNN) models for predicting the removal of As and other contaminants by iron-based filters from groundwater. A small Original Dataset (ODS) consisting of 20 data points and 13 groundwater parameters was obtained from the field performances of 20 individual iron-amended ceramic filters. Cubic-spline interpolation (CSI) expanded the ODS, generating 1600 interpolated data points (IDPs) without duplication. The Bayesian optimization algorithm tuned the model hyper-parameters and IDPs in a Stratified fivefold Cross-Validation (CV) setup trained all the models. The models demonstrated reliable performances with the coefficient of determination (R2) 0.990-0.999 for As, 0.774-0.976 for Iron (Fe), 0.934-0.954 for Phosphorus (P), and 0.878-0.998 for predicting manganese (Mn) in the effluent. Sobol sensitivity analysis revealed that As (total order index (ST) = 0.563), P (ST = 0.441), Eh (ST = 0.712), and Temp (ST = 0.371) are the most sensitive parameters for the removal of As, Fe, P, and Mn. The comprehensive approach, from data expansion through DLNN model development, provides a valuable tool for estimating optimal As removal conditions from groundwater.

Mapping global distributions of clay-size minerals via soil properties and machine learning techniques

Clay-size mineral is a vital ingredient of soil that influences various environment behaviors. It is crucial to establish a global distribution map of clay-size minerals to improve the recognition of environment variations. However, there is a huge gap of lacking some mineral contents in poorly accessible remote areas. In this work, machine learning (ML) approaches were conducted to predict the mineral contents and analyze their global abundance changes through the relationship between soil properties and mineral distributions. The average content of kaolinite, illite, smectite, vermiculite, chlorite, and feldspar were predicated to be 28.69 %, 22.30 %, 12.42 %, 5.43 %, 5.03 %, and 1.44 % respectively. Model interpretation showed that topsoil bulk density and drainage class were the most significant factors for predicting all six minerals. It could be seen from the feature importance analysis that bulk density notably reflected the distribution of 2:1 layered minerals more than that of 1:1 mineral. High drainage favored secondary minerals development, while low drainage was more benefited for primary minerals. Moreover, the content variation of different minerals aligned with the distribution of corresponding soil properties, which affirmed the accuracy of established models. This study proposed a new approach to predict mineral contents through soil properties, which filled a necessary step of understanding the geochemical cycles of soil-related processes.

Reclamation of co-pyrolyzed dredging sediment as soil cadmium and arsenic immobilization material: Immobilization efficiency, application safety, and underlying mechanisms

The safe management of toxic metal-polluted dredging sediment (DS) is imperative owing to its potential secondary hazards. Herein, the co-pyrolysis product (DS@BC) of polluted DS was creatively applied to immobilize soil Cd and As to achieve DS resource utilization, and the efficiency, safety, and mechanism were investigated. The results revealed that the DS@BC was more effective at reducing soil Cd bioavailability than the DS was (58.9-73.2% vs. 21.8-27.4%), except for the dilution effect, whereas the opposite phenomenon occurred for soil As (25.5-35.7% vs. 35.7-42.8%). The DS@BC immobilization efficiency was dose-dependent for both Cd and As. Soil labile Cd and As were transformed to more stable fractions after DS@BC immobilization. DS@BC immobilization inhibited the transfer of soil Cd and As to Brassica chinensis L. and did not cause excessive accumulation of other toxic metals in the plants. The appropriate addition of the DS@BC (8%) sufficiently alleviated the oxidative stress response of the plants and enhanced their growth. These findings indicate that the DS@BC was safe and effective for soil Cd and As immobilization. DS@BC immobilization decreased the diversity and richness of the rhizosphere soil bacterial community because of the dilution effect. The DS@BC immobilized soil Cd and As via direct adsorption, and indirect increasing soil pH, and regulating the abundance of specific beneficial bacteria (e.g., Bacillus). Therefore, the use of co-pyrolyzed DS as a soil Cd and As immobilization material is a promising resource utilization method for DS. Notably, to verify the long-term effects and safety of DS@BC immobilization, field trials should be conducted to explore the effectiveness and risk of harmful metal release from DS@BC immobilization under real-world conditions.

Adsorption behavior and mechanism of heavy metals onto microplastics: A meta-analysis assisted by machine learning

Microplastics (MPs) have the potential to adsorb heavy metals (HMs), resulting in a combined pollution threat in aquatic and terrestrial environments. However, due to the complexity of MP/HM properties and experimental conditions, research on the adsorption of HMs onto MPs often yields inconsistent findings. To address this issue, we conducted a comprehensive meta-analysis assisted with machine learning by analyzing a dataset comprising 3340 records from 134 references. The results indicated that polyamide (PA) (ES = -1.26) exhibited the highest adsorption capacity for commonly studied HMs (such as Pb, Cd, Cu, and Cr), which can be primarily attributed to the presence of C=O and N-H groups. In contrast, polyvinyl chloride (PVC) demonstrated a lower adsorption capacity, but the strongest adsorption strength resulting from the halogen atom on its surface. In terms of HMs, metal cations were more readily adsorbed by MPs compared with metalloids and metal oxyanions, with Pb (ES = -0.78) exhibiting the most significant adsorption. As the pH and temperature increased, the adsorption of HMs initially increased and subsequently decreased. Using a random forest model, we accurately predicted the adsorption capacity of MPs based on MP/HM properties and experimental conditions. The main factors affecting HM adsorption onto MPs were HM and MP concentrations, specific surface area of MP, and pH. Additionally, surface complexation and electrostatic interaction were the predominant mechanisms in the adsorption of Pb and Cd, with surface functional groups being the primary factors affecting the mechanism of MPs. These findings provide a quantitative summary of the interactions between MPs and HMs, contributing to our understanding of the environmental behavior and ecological risks associated with their correlation.

Predicting and refining acid modifications of biochar based on machine learning and bibliometric analysis: Specific surface area, average pore size, and total pore volume

Acid-modified biochar is a modified biochar material with convenient preparation, high specific surface area, and rich pore structure. It has great potential for application in the heavy metal remediation, soil amendments, and carrying catalysts. Specific surface area (SSA), average pore size (APS), and total pore volume (TPV) are the key properties that determine its adsorption capacity, reactivity, and water holding capacity, and an intensive study of these properties is essential to optimize the performance of biochar. But the complex interactions among the preparation conditions obstruct finding the optimal modification strategy. This study collected dataset through bibliometric analysis and used four typical machine learning models to predict the SSA, APS, and TPV of acid-modified biochar. The results showed that the extreme gradient boosting (XGB) was optimal for the test results (SSA R2 = 0.92, APS R2 = 0.87, TPV R2 = 0.96). The model interpretation revealed that the modification conditions were the major factors affecting SSA and TPV, and the pyrolysis conditions were the major factors affecting APS. Based on the XGB model, the modification conditions of biochar were optimized, which revealed the ideal preparation conditions for producing the optimal biochar (SSA = 727.02 m2/g, APS = 5.34 nm, TPV = 0.68 cm3/g). Moreover, the biochar produced under specific conditions verified the generalization ability of the XGB model (R2 = 0.99, RMSE = 12.355). This study provides guidance for optimizing the preparation strategy of acid-modified biochar and promotes its potentiality for industrial application.

Nano-enabled strategies to promote safe crop production in heavy metal(loid)-contaminated soil

Nanobiotechnology is a potentially safe and sustainable strategy for both agricultural production and soil remediation, yet the potential of nanomaterials (NMs) application to remediate heavy metal(loid)-contaminated soils is still unclear. A meta-analysis with approximately 6000 observations was conducted to quantify the effects of NMs on safe crop production in soils contaminated with heavy metal(loid) (HM), and a machine learning approach was used to identify the major contributing features. Applying NMs can elevate the crop shoot (18.2 %, 15.4-21.2 %) and grain biomass (30.7 %, 26.9-34.9 %), and decrease the shoot and grain HM concentration by 31.8 % (28.9-34.5 %) and 46.8 % (43.7-49.8 %), respectively. Iron-NMs showed a greater potential to inhibit crop HM uptake compared to other types of NMs. Our result further demonstrates that NMs application substantially reduces the potential health risk of HM in crop grains by human health risk assessment. The NMs-induced reduction in HM accumulation was associated with decreasing HM bioavailability, as well as increased soil pH and organic matter. A random forest model demonstrates that soil pH and total HM concentration are the two significant features affecting shoot HM accumulation. This analysis of the literature highlights the significant potential of NMs application in promoting safe agricultural production in HM-contaminated agricultural lands.

Improved mapping of heavy metals in agricultural soils using machine learning augmented with spatial regionalization indices

The accurate spatial mapping of heavy metal levels in agricultural soils is crucial for environmental management and food security. However, the inherent limitations of traditional interpolation methods and emerging machine-learning techniques restrict their spatial prediction accuracy. This study aimed to refine the spatial prediction of heavy metal distributions in Guangxi, China, by integrating machine learning models and spatial regionalization indices (SRIs). The results demonstrated that random forest (RF) models incorporating SRIs outperformed artificial neural network and support vector regression models, achieving R2 values exceeding 0.96 for eight heavy metals on the test data. Hierarchical clustering for feature selection further improved the model performance. The optimized RF models accurately predicted the heavy metal distributions in agricultural soils across the province, revealing higher levels in the central-western regions and lower levels in the north and south. Notably, the models identified that 25.78 % of agricultural soils constitute hotspots with multiple co-occurring heavy metals, and over 6.41 million people are exposed to excessive soil heavy metal levels. Our findings provide valuable insights for the development of targeted strategies for soil pollution control and agricultural soil management to safeguard food security and public health.

Automated machine learning-aided prediction and interpretation of gaseous by-products from the hydrothermal liquefaction of biomass

Hydrothermal liquefaction (HTL) is a thermochemical conversion technology that produces bio-oil from wet biomass without drying. However, by-product gases will inevitably be produced, and their formation is unclear. Therefore, an automated machine learning (AutoML) approach, automatically training without human intervention, was used to aid in predicting gaseous production and interpreting the formation mechanisms of four gases (CO2, CH4, CO, and H2). Specifically, four accurate optimal single-target models based on AutoML were developed with elemental compositions and HTL conditions as inputs for four gases. Herein, the gradient boosting machine (GBM) performed excellently with train R2 ≥ 0.99 and test R2 ≥ 0.80. Then, the screened GBM algorithm-based ML multi-target models (maximum average test R2 = 0.89 and RMSE = 0.39) were built to predict four gases simultaneously. Results indicated that biomass carbon, solid content, pressure, and biomass hydrogen were the top four factors for gas production from HTL of biomass. This study proposed an AutoML-aided prediction and interpretation framework, which could provide new insight for rapid prediction and revelation of gaseous compositions from the HTL process.

Application of machine learning approaches to predict ammonium nitrogen transport in different soil types and evaluate the contribution of control factors

The loss of nitrogen in soil damages the environment. Clarifying the mechanism of ammonium nitrogen (NH4+-N) transport in soil and increasing the fixation of NH4+-N after N application are effective methods for improving N use efficiency. However, the main factors are not easily identified because of the complicated transport and retardation factors in different soils. This study employed machine learning (ML) to identify the main influencing factors that contribute to the retardation factor (Rf) of NH4+-N in soil. First, NH4+-N transport in the soil was investigated using column experiments and a transport model. The Rf (1.29 - 17.42) was calculated and used as a proxy for the efficacy of NH4+-N transport. Second, the physicochemical parameters of the soil were determined and screened using lasso and ridge regressions as inputs for the ML model. Third, six machine learning models were evaluated: Adaptive Boosting, Extreme Gradient Boosting (XGB), Random Forest, Gradient Boosting Regression, Multilayer Perceptron, and Support Vector Regression. The optimal ML model of the XGB model with a low mean absolute error (0.81), mean squared error (0.50), and high test r2 (0.97) was obtained by random sampling and five-fold cross-validation. Finally, SHapely Additive exPlanations, entropy-based feature importance, and permutation characteristic importance were used for global interpretation. The cation exchange capacity (CEC), total organic carbon (TOC), and Kaolin had the greatest effects on NH4+-N transport in the soil. The accumulated local effect offered a fundamental insight: When CEC > 6 cmol+ kg-1, and TOC > 40 g kg-1, the maximum resistance to NH4+-N transport within the soil was observed. This study provides a novel approach for predicting the impact of the soil environment on NH4+-N transport and guiding the establishment of an early-warning system of nutrient loss.