Portable X-ray fluorescence (pXRF) spectrometers provide simple, rapid, nondestructive, and cost-effective analysis of the metal contents in soils. The current method for improving pXRF measurement accuracy is soil sample preparation, which inevitably consumes significant amounts of time. To eliminate the influence of sample preparation on PXRF measurements, this study evaluates the performance of pXRF measurements in the prediction of eight heavy metals’ contents through machine learning algorithm linear regression (LR) and multivariate adaptive regression spline (MARS) models. Soil samples were collected from five industrial sites and separated into high-value and low-value datasets with pXRF measurements above or below the background values. The results showed that for Cu and Cr, the MARS models were better than the LR models at prediction (the MARS-R² values were 0.88 and 0.78; the MARS-RPD values were 2.89 and 2.11). For the pXRF low-value dataset, the multivariate MARS models improved the pXRF measurement accuracy, with the R² values improved from 0.032 to 0.39 and the RPD values increased by 0.02 to 0.37. For the pXRF high-value dataset, the univariate MARS models predicted the content of Cu and Cr with less calculation. Our study reveals that machine learning methods can better predict the Cu and Cr of large samples from multiple contaminated sites.

Biochar, a porous carbonaceous material derived from the pyrolysis of biomass under oxygen-limited conditions, offers several advantages for environmental remediation, including a high specific surface area, ease of preparation, and abundant raw material sources. However, the application of pristine biochar is limited by its inherent physicochemical shortcomings, such as a lack of active functional groups and limited elemental compositions. To overcome these limitations, metal-modified biochars have garnered increasing attention. In particular, iron-manganese (Fe-Mn) modification significantly enhances the adsorption capacity, redox potential, and microbial activity of biochar, owing to the synergistic interactions between Fe and Mn. Iron-manganese-modified biochar (FM-BC) has demonstrated effective removal of heavy metals, organic matter, phosphate, and nitrate through mechanisms including mesoporous adsorption, redox reactions, complexation, electrostatic interactions, and precipitation. Moreover, FM-BC can improve soil physicochemical properties and support plant growth, highlighting its promising potential for broader environmental application. This review summarizes the preparation methods, environmental remediation mechanisms, and practical applications of FM-BC and discusses future directions in mechanism elucidation, biomass selection, and engineering implementation. Overall, FM-BC, with its tunable properties and multifunctional capabilities, emerges as a promising and efficient material for addressing complex environmental pollution challenges.

The widely used rubber antioxidant N-(1,3-Dimethylbutyl)-N'-phenyl-p-phenylenediamine (6PPD) and its ozonated product, 6PPD-quinone (6PPD-Q), are highly toxic to aquatic life, yet understanding on their environmental behaviors is limited. This study comprehensively investigated their adsorption and transformation processes on natural Fe-Mn nodules (NFMN), which commonly exist in sediments and soils through a combination of diverse experimental and computational methods. The maximum adsorption capacity of 6PPD-Q (719.2 μg·g^-1) is significantly higher than that of 6PPD (133.8 μg·g^-1) at 293 K, and it is more difficult to desorb. They follow different kinetic and isothermal adsorption models, and environmental conditions (including temperature, pH, and anions) exert distinct influences on the adsorption of the two substances. Adsorption mechanisms involving electrostatic attraction, charge transfer, hydrogen bonding, and Lewis acid-base complexation were unveiled. For 6PPD, electrostatic adsorption and Lewis acid-base complexation contribute significantly to its adsorption. Conversely, for 6PPD-Q, the contribution of Lewis acid-base complexation outweighs that of hydrogen bonding, while the effect of electrostatic adsorption is relatively negligible. The stronger electrostatic attraction, more efficient charge transfer, and a greater number of binding sites for hydrogen bonding and Lewis acid-base complexation with NFMN results in more robust adsorption of 6PPD-Q. Furthermore, 6PPD can transform into 6PPD-Q on NFMN, facilitated by dissolved Mn(III). This study advances understanding of the adsorption behavior and mechanism of 6PPD and 6PPD-Q, and highlights a new pathway for 6PPD-Q formation, which provides valuable reference for assessing the water body exposure risks and formulating environmental remediation strategies for such pollutants. ENVIRONMENTAL IMPLICATION: This study offers the first comprehensive insight into the interactions between 6PPD/6PPD-Q and NFMN, illuminating their environmental behavior in water and soil systems. It reveals the adsorption discrepancy between 6PPD and 6PPD-Q, and elucidates the mechanisms underlying the difference in adsorption. Additionally, it uncovers a novel pathway for 6PPD-Q formation, offering valuable implications for risk assessment and environmental remediation strategies.

Chlorinated aliphatic hydrocarbons (CAHs) are widely used in agriculture and industries and have become one of the most common groundwater contaminations. With the excellent performance of the deep learning method in predicting, LSTM and XGBoost were used to forecast dichloroethene (DCE) concentrations in a pesticide-contaminated site undergoing natural attenuation. The input variables included BTEX, vinyl chloride (VC), and five water quality indicators. In this study, the predictive performances of long short-term memory (LSTM) and extreme gradient boosting (XGBoost) were compared, and the influences of variables on models’ performances were evaluated. The results indicated XGBoost was more likely to capture DCE variation and was robust in high values, while the LSTM model presented better accuracy for all wells. The well with higher DCE concentrations would lower the model’s accuracy, and its influence was more evident in XGBoost than LSTM. The explanation of the SHapley Additive exPlanations (SHAP) value of each variable indicated high consistency with the rules of biodegradation in the real environment. LSTM and XGBoost could predict DCE concentrations through only using water quality variables, and LSTM performed better than XGBoost.