（文章来源：科技日报） 韩国蔚山科学技术院（UNIST）研究团队成功研制出一种能在"柔软灵活"与"坚硬有力"状态间切换的新 型人造肌肉。实验表明，其仅重1.25克却性能卓越：刚性时可承载5千克重物（约为自身重量的4000 倍），柔性时伸展率达原始长度12倍。相关成果近日发表于《先进功能材料》杂志。 ...

Core Viewpoint - The announcement by China Unicom, China Mobile, and China Telecom regarding the approval for eSIM mobile service trials marks a significant step in the commercial deployment of eSIM technology in China, transitioning from IoT and wearable devices to mobile phones [1] Group 1: eSIM Technology Overview - eSIM, or embedded SIM, integrates the SIM card directly into the device's chip, eliminating the need for a physical SIM card [2] - eSIM technology allows for multiple SIM profiles, providing greater internal space for other components and enhancing waterproof and dustproof capabilities due to the absence of a SIM card slot [2] Group 2: Security Aspects of eSIM - eSIM reduces the risks of loss, theft, or tampering compared to traditional SIM cards, as it is embedded in the device and requires physical damage to remove [3] - The security of eSIM is enhanced through encryption protocols and requires user authorization for activation, making it more difficult to clone or hack compared to traditional SIM cards [3] Group 3: User Experience and Business Operations - Users must visit physical stores to activate eSIM services, as online activation is not yet available [4] - Users can convert existing physical SIM cards to eSIM and manage their eSIM profiles through service providers, with specific procedures for deactivating or transferring eSIM numbers [4]

科技日报北京10月14日电 （记者付毅飞）记者14日从中国载人航天工程办公室获悉，由全国载人 航天标准化技术委员会组织提出的国际标准项目ISO/NP14620-5《航天系统—安全性要求—第5部分：载 人航天器》，近日在国际标准化组织（ISO）正式注册立项。 据悉，这是我国载人航天领域首个成功立项的国际标准，也是该领域在国际规则制定中的首次突 破，为促进载人航天飞行安全提供了"中国方案"，将进一步提升我国载人航天国际话语权和影响力。 后续，项目组将按计划开展各阶段编制工作，确保标准顺利发布、实施。 该标准内容包括载人航天器安全性管理要求、安全性技术要求、安全性设计与验证要求、在轨飞行 任务安全性要求等，从保证人的生命安全角度，从顶层规范了载人航天任务研制、发射、在轨运行、返 回、着陆等阶段的安全性工作，适用于载人飞船、货运飞船、空间站以及相关载荷产品等，对于月球探 测工程相关产品安全性设计也具有指导意义。 该标准在研究和借鉴国外航天领域安全性先进技术和方法的基础上，总结我国载人航天工程30余年 安全性工作经验，提出适用于各国载人航天器及相关产品开展安全性工作的统一技术指导和范式，向全 世界公布参与载人航天工作的 ...

Core Insights - The CRAFT project, led by the Hefei Institute of Physical Science, has achieved significant progress with the successful testing and acceptance of the prototype component for the divertor, which is crucial for fusion reactors [1][2] - The prototype's steady-state thermal load capacity reached 20 megawatts per square meter, and the alignment error of the target plate surface to the plasma is less than 1 millimeter, marking it as the largest and highest thermal load divertor prototype designed in China [1] - The innovative design of the hybrid divertor cladding integration is expected to enhance the tritium breeding ratio by over 3%, providing an effective means for achieving tritium self-sufficiency [1] Technical Developments - The divertor prototype is a key component among 19 systems in the CRAFT project, indicating that China's divertor research and development capabilities have become self-controlled [2] - The successful development of this component lays a solid technical foundation for future engineering applications of divertors in Chinese fusion reactors [2] - The technology developed can also support other fusion devices facing plasma components and has potential applications in aerospace, high-end medical equipment, industrial electronics, and new energy vehicles [2]

Core Insights - The integration of AI in materials research is seen as a transformative force, with significant advancements made by companies like DeepMind and Microsoft in discovering new materials [1][2][3] - Despite the enthusiasm, there are criticisms regarding the originality and practicality of AI-generated compounds, highlighting the need for collaboration with experimental chemists [1][4][6] Group 1: AI Advancements in Materials Research - DeepMind's GNoME AI system discovered 2.2 million new crystal materials, including 52,000 graphene-like compounds and 528 lithium-ion conductors [2] - The A-Lab robotic system by Lawrence Berkeley National Laboratory synthesizes compounds predicted by DFT, demonstrating the ability to create previously unmade materials [2] - Microsoft's MatterGen tool generates materials based on specified mechanical, electrical, and magnetic properties, enhancing targeted research [3] Group 2: Criticism and Controversies - Critics argue that some AI-generated compounds lack originality and practical value, with examples of rare radioactive elements being included in predictions [4] - A-Lab's results faced scrutiny, with claims of inaccuracies in synthesized compounds, although the lab defended its findings [4][5] - MatterGen's recommendations included materials that were already known, raising questions about the novelty of its outputs [4] Group 3: Future Directions and Challenges - Most researchers believe that with continuous optimization, AI models can significantly advance materials science [6] - Microsoft is developing MatterSim to validate the stability of structures proposed by MatterGen under real conditions, addressing reliability concerns [7] - The demand for new materials to tackle societal challenges is driving ongoing exploration of AI in this field, with companies like Citrine Informatics customizing AI systems to enhance material optimization [7]

Core Insights - Researchers at Cambridge University have developed "blood-like cells" using human stem cells, which can simulate key stages of early human development, including the generation of blood stem cells [1][2] - The new human embryo-like model accurately replicates the initiation of the hematopoietic system in embryos, providing a powerful tool for drug screening, early blood and immune system development research, and modeling blood diseases [2][3] Group 1 - The three-dimensional structures created by human stem cells exhibit self-organization capabilities, forming the three primary germ layers (ectoderm, mesoderm, and endoderm) within two days of culture [1] - By day eight, beating heart cells were observed, which in real embryos develop into the heart, and by day thirteen, functional blood cells were confirmed with visible red blood spots [1][2] - The ability to produce human blood cells in the lab marks a significant step in regenerative medicine, allowing for the potential creation of genetically matched blood cells for patients, thus avoiding immune rejection [2] Group 2 - The model captures the "second wave" of hematopoiesis during human development, which includes the production of adaptive lymphocytes such as T cells, opening new avenues for studying blood development in both healthy and cancerous states [2] - This research adheres to international ethical standards and has received approval from ethics committees, ensuring compliance with regulatory frameworks [1] - The technology may eventually provide tailored blood cells or hematopoietic stem cells for patients with blood diseases like leukemia, potentially saving more lives [3]

Core Insights - A research team from Peking University has developed a high-precision, scalable analog matrix computing chip based on resistive random-access memory (ReRAM), achieving analog computing precision comparable to digital systems [1][2] - The chip significantly enhances computational throughput and energy efficiency, with improvements ranging from 100 to 1000 times over current top digital processors (GPUs) [1][2] Group 1 - The chip addresses complex matrix equation solving, which is essential for applications like communication base station signal processing and AI model training [1] - The research team utilized a novel approach combining new information devices, original circuits, and classical algorithms to create a full analog matrix equation solver with 24-bit fixed-point precision [1][2] - The team successfully demonstrated a relative error as low as 10^-7 after 10 iterations for a 16x16 matrix inversion, showcasing the chip's high precision [2] Group 2 - In terms of performance, the chip surpasses high-end GPU single-core performance when solving 32x32 matrix inversion problems, achieving over 1000 times the throughput of traditional digital processors for 128x128 matrices [2] - The chip's energy efficiency is over 100 times better than traditional digital processors, making it a critical technology for high-efficiency computing centers [2] - The application of this technology in large-scale MIMO signal detection demonstrated high fidelity in image recovery with a bit error rate comparable to 32-bit digital calculations, highlighting its potential in real-time signal processing [2] Group 3 - The breakthrough in analog computing is expected to reshape the computational landscape, providing a promising path for enhancing computational power and potentially breaking the long-standing dominance of digital computing [3]

美国科学家研发出一种形似"中国灯笼"的聚合物结构，通过压缩或扭曲等简单操作，就能快速变换出十 几种不同的三维弯曲形态。更引人瞩目的是，这种变形过程可通过磁场远程操控，大大拓展了其应用潜 力。相关研究成果发表于最新一期《自然·材料》杂志。 研究还发现，通过扭曲结构、将灯笼顶部或底部的实心条向内或外折叠，或是组合使用这些操作，还能 创造出更多形态。 通过在结构底部实心条上附着磁性薄膜，研究团队实现了利用磁场远程控制结构的压缩与扭曲。他们还 演示了基于双稳态切换的多种应用场景，例如非侵入式鱼类捕捉器、可调控水流开合的过滤器等。 这类"灯笼"结构未来有望组装成二维或三维系统，在可变形的机械超材料与机器人技术领域大展身手。 （文章来源：科技日报） 这项研究由美国北卡罗来纳州立大学与宾夕法尼亚大学的科研团队合作完成。研究团队先将聚合物薄片 切割成钻石状的平行四边形，再于每条边中央刻出一排平行缝隙，从而在薄片上下两侧形成由实心材料 条连接的均匀色带。接着，将上下实心条的左右两端相连，便构成了一个近似"中国灯笼"的三维结构。 研究团队表示，这一基础结构本身具有"双稳态"特性。若从顶部向下压缩结构，它会逐渐变形，直至达 到某个 ...

Core Insights - Researchers at Cambridge University have developed "blood-like cells" using human stem cells, which can simulate multiple key stages of early human development, including the generation of blood stem cells [1][2] - The new human embryo-like model accurately replicates the initiation of the hematopoietic system in embryos, providing a powerful tool for drug screening, early blood and immune system development research, and modeling blood diseases [2] Group 1 - The embryo-like structures exhibit self-organization capabilities, forming the three primary germ layers (ectoderm, mesoderm, and endoderm) by day two of cultivation [1] - By day eight, beating heart cells were observed, which in real embryos will eventually develop into the heart [1] - On day thirteen, the team noted distinct red blood spots, confirming the generation of functional blood cells [1] Group 2 - The ability to produce human blood cells in the laboratory marks a significant step in regenerative medicine, allowing for the potential creation of blood cells that are genetically matched to patients, thus avoiding immune rejection [2] - The model captures the "second wave" of hematopoiesis during human development, which can produce adaptive lymphocytes, including T cells, opening new avenues for studying blood development in both healthy and cancerous states [2]

日本冲绳科学技术大学院大学科学家成功研制出一种近乎零摩擦的可自由悬浮旋转的石墨转子。这一成 果解决了长期困扰宏观悬浮系统的"涡流阻尼"问题，为高精度测量和量子研究开辟了新途径。相关论文 发表于新一期《通讯·物理》杂志。 最新设计在理想条件下彻底消除了涡流阻尼。团队表示，改进后的自由悬浮转子有望成为毫米级乃至微 米级高精度传感器核心部件，可用于高灵敏度陀螺仪，也可在低温下进入量子态，用于探索真空引力效 应等宏观量子现象。 （文章来源：科技日报） 与依赖复杂光学或电学系统的微尺度装置不同，宏观系统可在常温下实现磁悬浮，结构更简单、环境适 应性更强。然而，这类系统长期受涡流阻尼限制。导体在非均匀磁场中运动时会产生环状电流，形成抵 抗运动的磁场，类似"磁性摩擦"，导致能量损耗和信号干扰。 此次，团队利用直径约1厘米的石墨圆盘和稀土磁体，设计出抗磁悬浮转子。通过实验和数学分析证 明，只要系统具备完美轴对称性，就能完全消除涡流阻尼。如果能进一步减缓旋转速度，转子运动将进 入量子态，为宏观量子研究提供新平台。 人们看到物体飘浮在空中会惊叹不已，而科学家则将这种"失重"状态视为隔绝外界干扰的一种理想方 式。对科学家而言，悬浮 ...