A study in Forest Ecosystems shows that aquaculture wastewater may weaken carbon stabilization and alter nitrogen regulation in mangrove soils. In China’s Maowei Sea, heavily affected soils contained much less dissolved organic carbon and substantially less organic carbon associated with crystalline iron minerals. Nutrient enrichment appeared to stimulate microbial carbon consumption and more reducing soil conditions, weakening iron-organic carbon coupling. At the same time, nitrogen cycling shifted toward greater ammonium retention and recycling rather than permanent removal. These changes could reduce the capacity of mangrove soils to preserve carbon over the long term and remove excess nitrogen, highlighting the importance of managing aquaculture discharge and monitoring carbon–iron-microbial interactions.

Weak polarization electric field (PEF) III-nitride LEDs offer a promising route toward high-performance micro-LEDs. To realize this promise, researchers from Jilin University and KAUST demonstrate record-high efficiency weak-PEF LEDs on polar c-plane, enabled by optimized MOCVD pulse-growth mode. The devices achieve the highest peak EQE of their kind, great wavelength stability, and strong lateral carrier confinement. This work represents a meaningful step toward micro-LED applications where efficiency, wavelength stability, and lateral carrier confinement are critical.

A new study finds that socio-culturally responsive virtual reality (VR) training can strengthen English teachers’ pragmatic competence—the ability to use language appropriately across social and cultural contexts. Involving 122 EFL teachers, the 108-hour intervention used fully immersive simulations featuring AI-driven avatars. Participants showed statistically significant improvements in pragmalinguistic accuracy and sociopragmatic appropriateness. Focus-group findings highlighted immersive engagement, adaptive feedback, communicative confidence, metapragmatic awareness, and reflection as important contributors to the observed improvements.

Researchers have developed a multimodal artificial intelligence framework that predicts the research octane number (RON) of both pure fuel components and complex mixtures with high accuracy. For pure components, the model achieved an R² of 0.9373 and a mean absolute error (MAE) of 4.04. For mixtures, the latent-space model achieved an R² of 0.9736 and an MAE of 1.46—reducing prediction error by 69.8 % compared with conventional linear blending rules. The model also enables inverse fuel formulation design, allowing researchers to search for feasible component proportions that meet target RON constraints.

Researchers have established a molecular‑scale reaction kinetic model for n‑hexane catalytic cracking over HZSM‑5 zeolite, revealing that isohexane is an important primary product and that the initial isomerization of n‑hexane to isohexane plays a critical role in the reaction network. Comparative analysis of reaction networks with and without isohexane showed that isomerization directly influences ethylene and propylene formation pathways and indirectly affects substances like isobutane. The kinetic model incorporating isomerization reduced the mean absolute error from 1.13 wt% to 0.21 wt%. Co‑feeding experiments confirmed that adding isohexane boosts ethylene selectivity and suppresses propylene formation.

Researchers have discovered that simply maximising the surface area of catalyst supports does not guarantee better performance. In a study on catalysts for producing a key biodegradable plastic precursor, a catalyst with the highest surface area performed the worst. The team identified two factors—the density of surface silanol (Si–OH) groups and pore size—as being far more important. By optimising these features, they achieved 92.2 % dimethyl oxalate conversion and 94.8 % methyl glycolate selectivity at 220 °C, offering a more practical route for manufacturing biodegradable polyglycolic acid.