Published chemical literature has accumulated a wealth of reaction data, but this data comes from diverse sources, is unevenly distributed, and is clearly biased towards highly selective "successful experiments." Can such literature data be used for reliable machine learning modeling and further guide new catalytic reactions? Professor Luo Sanzhong's team and collaborators at Tsinghua University recently reported OrgAIcat, a machine learning framework for asymmetric organocatalysis. The research team manually compiled over 22,000 data points on Aldol and Michael addition reactions from 844 papers to construct the iSynth dataset. Based on this, the team developed R-SPOC, a reaction descriptor that requires no quantum chemical calculations, for predicting enantioselectivity and screening catalysts. When OrgAIcat was combined with Bayesian optimization, a challenging Aldol reaction was improved from 41% ee to 96% ee, with the entire optimization process requiring only 12 experiments. These results demonstrate that after rigorous compilation, proper characterization, and sufficient validation, historical literature data can still be used to guide experiments, despite its heterogeneity and publication bias. The article was published as an open access Research Article in CCS Chemistry, the flagship journal of the Chinese Chemical Society.

Valleytronics, which uses the valley degree of freedom of electrons to encode information, offers a promising route toward next-generation low-power electronic and optoelectronic devices. Researchers from Xiamen University have now shown that introducing graphene into a WS₂/CrBr₃ van der Waals heterostructure can dramatically accelerate interlayer spin-charge transfer, leading to giant valley polarization splitting and markedly improved circularly polarized light detection.

The design of unique fluorescent probes is of paramount importance for boosting the signal-to-noise ratio (SNR) and improving the sensitivity toward target analytes. Herein, researchers from the Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences have reported a hydrogen-bond-driven self-assembled sensing system (CdSe/ZnS QDs@FNAC) composed of quantum dots (QDs) capped with 3-mercaptopropionic acid (CdSe/ZnS QDs@MPA) and a naphthol-based molecular probe (FNAC). This system enables a ratiometric fluorescent dual-signal response toward KMnO₄ based on the synergistic mechanism of photoinduced electron transfer (PET) and twisted intramolecular charge transfer (TICT), featuring excellent sensing performance including a fast response speed (1 s), high sensitivity (4.80 nM), a favorable SNR, and negligible interference from other oxidants. The practicality of CdSe/ZnS QDs@FNAC was further verified by integrating a CdSe/ZnS QDs@FNAC-embedded hydrogel-based sensing chip into a portable detector, which enabled the on-site identification of ng-level KMnO₄ particles with high selectivity.

With the rapid development of electronic communication, aerospace, military and medical equipment, the demand for high‑performance electromagnetic interference (EMI) shielding materials has grown dramatically. A review article led by Prof. Ming‑Guo Ma from Beijing Forestry University systematically summarizes the latest progress in MXene/nanocellulose composites for EMI shielding. The work focuses on three key aspects: preparation strategies, toughening mechanisms (physical, chemical and synergistic), and multidimensional structural design (1D fibers, 2D layered films and 3D networks). The review also analyses current technological limitations and proposes future research directions, offering a roadmap for the rational design of high‑performance, flexible and lightweight EMI shielding composites.

A key challenge in transdermal delivery is maintaining consistent performance under dynamic conditions created by everyday user activities, such as skin bending, stretching, and movement. Addressing these real-world conditions is essential for both reliable drug delivery and user convenience. The bioinspired suction-activated nanofiber patch developed in this study achieves robust skin adhesion and consistent, enhanced transdermal delivery even under dynamic skin environments. By combining strong interfacial stability with user-friendly applications, this platform highlights a promising new direction for next-generation transdermal drug delivery systems.