Plastic waste should not be viewed as a uniform refuse stream, but as a heterogeneous carbon reservoir whose recycling pathway is dictated by polymer composition, contamination level and degradation history. Here, we outline a graded recycling framework that prioritizes carbon retention rather than technology-driven conversion. Clean and separable plastic streams should first undergo physical recycling to preserve polymer backbones and avoid unnecessary bond cleavage. Mixed, aged or contaminated fractions are better suited to selective chemical routes, including solvolysis, hydrogenolysis, catalytic depolymerization, electrified pyrolysis and gasification, which can reconstruct carbon into monomers, hydrocarbons, syngas or platform chemicals. Biological recycling is most effective for oxygenated intermediates generated through pretreatment. Integrating sorting, safety control and life-cycle assessment, this physical–chemical–biological hierarchy matches each plastic carbon stream with its most valuable and sustainable conversion pathway.

Ultra-high temperature ceramics (UHTCs) are critical to thermal protection systems for hypersonic vehicles, yet their oxidation behavior above 2000 ℃ remains poorly understood because conventional testing methods have limited applicability. Researchers at Donghua University, China, have developed a novel ultra-fast heating method (100 ℃/s, >2000 ℃) to evaluate the cyclic oxidation behavior of multi-principal component diboride (MEB) ceramics at 2300 ℃. Among the compositions studied, (Zr_1/4Hf_1/4Ta_1/4Ti_1/4)B₂ exhibited the best performance, forming a dense oxide scale just 99 mm thick that remained intact after five thermal cycles. The team revealed that beyond configurational entropy, the functional element Ti plays a dual role—densifying the outer oxide layer at high temperatures and inducing a stress-relieving porous needle-like Ta₂O₅ network during cooling. This work establishes a convenient and efficient testing protocol and provides a “structural-functional element synergy” design principle for reusable thermal protection materials.

This review summarizes recent progress in microwave-driven catalytic conversion of waste plastics. It covers the synergy between microwave dielectric heating and catalysis, depolymerization pathways of various plastics, catalyst design and structure-activity relationships, and effects of reaction parameters on product distribution. Key challenges are discussed, including unclear mechanisms, catalyst deactivation, and scale-up difficulties. Future directions are proposed, such as multi-scale mechanistic studies, development of stable catalysts, and continuous-flow reactor design.

MAX phase ceramics are promising candidates for high-temperature structural applications, yet the synthesis of o-MAX phases with out‑of‑plane chemical ordering is constrained by strict elemental compatibility. A research team led by Associate Professors Cheng-Feng Du and Hong Yu from Northwestern Polytechnical University recently reported a series of Cr₂(Ti_1-xV_x)AlC₂ (x = 0–0.6) quinary o-MAX materials. They established the V solubility threshold at x = 0.5, beyond which an order-disorder transition occurs. The optimal composition achieves 16.4% higher nanoindentation hardness, 35.2% higher Vickers hardness, and 5.1% higher fracture toughness, demonstrating that site-selective solid solution is a viable route to high-performance MAX ceramics.

Methane dry reforming converts CO₂ and CH₄ into syngas for green fuel synthesis to advance carbon neutrality. Low-cost Ni catalysts suffer severe deactivation from sintering and coking.We fabricate a robust Ni_BN catalyst with uniformly dispersed 6 nm Ni nanoparticles withelectron-deficient Ni speciestailored by interfacial boron, delivering coke-free operation over 350 hours. This unique structure suppresses deep CH₄ cleavage while boosts CO₂ reactivity. It offers new guidance for C–H activation related catalyst design.

Journal of Integrative Medicine (JIM), an international English-language peer-reviewed journal focusing on traditional Chinese medicine (TCM), integrative medicine and related interdisciplinary research, has joined EurekAlert! to further strengthen the international dissemination of research in the field. Founded in 2013, JIM publishes high-quality basic, clinical and evidence-based research and is indexed in major databases including SCIE, MEDLINE/PubMed and Scopus. With a 2025 Journal Impact Factor of 5.2 and a JCR Q1 ranking, JIM aims to provide an international platform for advancing scientific exchange and sharing innovative research in integrative medicine.

The review was led by Prof. Shaorong Gao of Tongji University, in collaboration with researchers from multiple universities and research institutes across China. It examines how epigenetic mechanisms decide which genes are used, when they are used and how strongly they are expressed - from the earliest stages of development to aging, cancer and therapy.

Bringing together evidence from reproduction, embryonic development, stem-cell research, tissue maintenance, aging, cancer and clinical translation, the review assesses what these mechanisms may mean for reproductive medicine, regenerative medicine and precision treatment. It also distinguishes promising directions from interventions that still require stronger evidence on delivery, specificity and long-term safety.

In a paper published in SCIENCE CHINA Earth Sciences, researchers at Sun Yat-sen University developed an SFA-based Climate Prediction (SCP) method, which directly extracted slow-varying predictable signals from observed precipitation data and applied it to seasonal prediction of summer precipitation over South China. Independent forecasts for 2011–2022 show that the SCP method achieves a regional mean temporal correlation coefficient of 0.36 and accurately predicted the sign of regional-mean precipitation anomalies in 10 of 12 summers, outperforming two dynamical models and a hybrid dynamical–statistical prediction system. The study provides a new perspective to identifying potentially predictable signals and improving regional seasonal precipitation prediction.

Robots often learn from historical experience collected under older controllers and physical conditions. When tasks, hardware or environments change, that data can become misleading. Researchers developed Occupancy-Matching Policy Optimization (OMPO), which anchors large replay buffers to a small set of recent interactions. Across locomotion, transfer, manipulation and a real-robot test, OMPO improved stability, sample efficiency and adaptation.

A recent review outlines major advances in hydrogel-based strategies for musculoskeletal therapy. Through refined material design and strengthened mechanical performance, hydrogels can mimic key extracellular matrix features while enabling localized and controlled delivery of drugs, cells, and other bioactive cues. They can also be engineered into clinically relevant formats, including injectable systems, scaffolds, patches, and adhesive sealants, supporting applications across bone and cartilage repair, osteoarthritis management, disc regeneration, tendon healing, and dural repair.