Bochum, Germany, October 29, 2025, Researchers from Research Center for Future Energy Materials and Systems at the Ruhr University Bochum, Software for Chemistry & Materials BV, and Vrije Universiteit Amsterdam have demonstrated that modern universal machine learning interatomic potentials (uMLIPs) can now accurately describe systems ranging from single molecules to bulk solids, representing a significant leap forward for uMLIPs in materials science. The study introduces the 0123D dataset, comprising 40,000 diverse structures specifically designed to benchmark model performance across all dimensionalities.

In a study published in Advanced Materials, a research team led by Academician YU Shuhong from the University of Science and Technology of China (USTC) of the Chinese Academy of Sciences reported a structure-function integrated design of nacre-mimetic alumina (Al2O3)-based (NMA) composites. This NMA composite material not only has unique color tunability and excellent wave transparency, but also achieves lightweight, high strength, high toughness, and outstanding impact resistance.

A research team from Yunnan University has developed a novel liquid metal-assisted heteroepitaxy method to grow high-quality perovskite crystals within mesoporous scaffolds. This breakthrough enables printable mesoscopic perovskite solar cells to reach a champion efficiency of 20.2% while maintaining 97% performance after 3000 hours under harsh conditions. The approach offers a scalable pathway to efficient, stable, and low-cost printable photovoltaics.

This study provides structural insights that inform the design and development of next-gen antidiabetic drug.

In a paper published in SCIENCE CHINA Earth Sciences, a team of researchers investigated a fine-scale lightning forecasting approach based on weather foundation models (WFMs) and proposed a dual-source data-driven forecasting framework that integrates the strengths of both WFMs and recent lightning observations to enhance predictive performance. Furthermore, a gated spatiotemporal fusion network (gSTFNet) is designed to address the challenges of cross-temporal and cross-modal fusion inherent in dual-source data integration. Experimental results demonstrate that the dual-source framework significantly improves forecasting performance compared to models trained solely on WFMs and outperforms both the ECMWF HRES lightning product and other deep-learning spatiotemporal forecasting models.

New study reveals leafy crops accumulate more resistance genes than fruiting plants, offering insight into safer agricultural practices.

Two unique hallmarks in non-Hermitian systems, namely exceptional points (EPs) and the non-Hermitian skin effect (NHSE), have attracted considerable attention, particularly regarding their interplay. Recently, researchers in China demonstrated that coupling two non-Hermitian systems, each exhibiting independent NHSE, can induce multiple pairs of EPs. This further realizes the manipulation of wave-function localization, which is experimentally characterized on integrated photonic platforms. This work may inspire new designs for photonic integrated devices based on non-Hermitian physics.