7-Apr-2026 Machine learning reveals how biochar can curb phosphorus pollution from farmland Biochar Editorial Office, Shenyang Agricultural University A new study has shown that biochar, a carbon-rich material produced from biomass, can significantly reduce phosphorus losses from agricultural soils, offering a promising solution to one of the leading causes of water pollution worldwide. By combining environmental science with machine learning, researchers have developed a powerful predictive tool to optimize how biochar is used in real-world farming systems. Journal Biochar
7-Apr-2026 Engineered biochar and bacteria team up to lock toxic metals in polluted soils Biochar Editorial Office, Shenyang Agricultural University Heavy metal contamination from industrial activities remains a major environmental challenge worldwide, especially at sites affected by lead and zinc smelting. Now, a new study offers a promising, low-cost strategy to stabilize multiple toxic metals in soil using a novel combination of biochar and beneficial bacteria. Journal Biochar
7-Apr-2026 Light-driven method enables sustainable production of porous semiconducting polymers Koç University A new study reports a visible-light-driven method to synthesize porous semiconducting polymers under ambient conditions without metal catalysts, offering a more sustainable and efficient route with strong photocatalytic performance. Journal Nature Communications
7-Apr-2026 Machine-learning guides discovery of multi-principal element alloys as electrocatalyst for hydrogen evolution reaction HEP Data Cooperation Journals This study develops an ML-assisted approach for accelerating the discovery of HER electrocatalysts in multi-principal element alloys. With ultralow material costs and record-breaking electrocatalytic metrics, the NbZnCo2 alloy exhibits exceptional potential for industrial alkaline water electrolyzers. Journal Acta Physico-Chimica Sinica
7-Apr-2026 Machine learning potentials for property predictions of two-dimensional group-III nitrides HEP Data Cooperation Journals A high-precision machine learning potential based on the Deep Potential method was constructed to systematically investigate the lattice dynamics, thermodynamic properties, mechanical behaviors, and thermal transport of two-dimensional Group-III nitrides with first-principles accuracy. Journal Acta Physico-Chimica Sinica
7-Apr-2026 LASPAI: AI-powered platform for the future atomic simulation HEP Data Cooperation Journals A web-based platform integrating generalized global neural network potentials and diffusion generative models was developed to resolve the accuracy-efficiency dilemma and democratize access to complex atomic simulations. Journal Acta Physico-Chimica Sinica
7-Apr-2026 High-rate and long-cycling P2-type cathode material for sodium-ion batteries HEP Data Cooperation Journals A multi-element doping strategy was employed to design a stable P2-type cathode (P2-Na0.67Zn0.05Ni0.23Fe0.1Mn0.57Ti0.05O2) that suppresses high-voltage phase transition, demonstrating a capacity retention of over 85% after 300 cycles at a 3C rate. Journal Acta Physico-Chimica Sinica
7-Apr-2026 Data-driven framework unlocks fast track to high-performance MnO2 cathodes HEP Data Cooperation Journals This study develops a high-accuracy machine-learning framework to predict and optimize metal-doped MnO2 cathodes for aqueous zinc-ion batteries, validated through experiments and DFT, enabling rapid performance guidance. Journal Acta Physico-Chimica Sinica
7-Apr-2026 CADD and AI for the next-generation therapeutics HEP Data Cooperation Journals The deep integration of computer-aided drug discovery (CADD) and artificial intelligence (AI) is ushering in a new era of precision and efficiency in next-generation therapeutics by reshaping R&D paradigms, unlocking the potential of biological big data, and accelerating the development of innovative therapies. Journal Pharmaceutical Science Advances
