Quantum entanglement stands as an intriguing and pivotal phenomenon in quantum mechanics, forming the cornerstone for the realization of technologies transcending classical boundaries and enabling the practical deployment of diverse quantum devices. Here, we propose a method to manipulate both the pairwise entanglement among exciton-photon, exciton-phonon, and photon-phonon modes, and the tripartite entanglement involving exciton, photon, and phonon modes, within a hybrid exciton-optomechanics system, utilizing two pump lasers propagating in different directions. We find that all three distinct types of bipartite quantum entanglement and the tripartite entanglement can be effectively controlled and even enhanced along with improved robustness against environmental thermal noise by properly adjusting the phase difference between the two pump lasers. Our findings open up an easily accessible method for manipulating and protecting quantum entanglement, which is crucial in a wide range of practical applications requiring purely quantum resources, such as noise-tolerant quantum processing and quantum precision measurements.
The uncontrollable growth of lithium dendrites has been regarded as a critical challenge to the advancement of high-energy lithium-metal batteries (LMBs) for future multi-scenario applications. Emerging Janus/gradient anode structures represent a transformative advancement in addressing the persistent challenges of lithium dendrite growth and volume fluctuation in LMBs. This review systematically summarizes the design principles, operational mechanisms, and recent progress of Janus and gradient anode architectures, highlighting their innovative role in guiding uniform lithium deposition and enhancing interfacial stability. Unlike conventional 3D current collectors, these structures feature asymmetric or gradient properties—such as lithiophilicity, conductivity, or porosity—enabling bottom-up plating and efficient space utilization. Key topics include lithiophilic-lithiophobic Janus designs, conductivity-gradient frameworks, and dual-gradient configurations, which collectively contribute to improved Coulombic efficiency, cyclic longevity, and safety. The review concludes with a perspective on future research directions, underscoring the potential of Janus/gradient anodes in enabling high-energy-density and durable LMBs.
Effect of acid impregnation and mechanical milling on the hydrolysis behavior and products distribution of chitin in subcritical water was investigated by a “temperature-programmed hydrolysis process”, in the present study. The acid impregnation and mechanical milling efficiently destroyed the glycosidic and hydrogen bonds in chitin, leading to a loss of crystal structure and significantly increased water solubility. The loss of crystal structure rendered the pretreated chitin substrates more susceptible to hydrolysis in subcritical water compared to raw chitin, thereby reducing the activation energy required to initiate the hydrolysis reaction. The main compounds in hydrolysate were analyzed to be reducing sugar, antioxidant compounds, 5-hydroxymethylfurfural, isopropyl acetate, acetamide and propanoic acid ethyl ester. The possible mechanism of acid impregnation and ball milling for hydrolysis of chitin in subcritical water was also proposed. This work suggested that acid-mechanical milling is an efficient pretreatment process for accelerating the liquefaction of chitin to valuable chemicals in subcritical water.
Current AI agents can flexibly invoke tools and execute complex tasks, yet their long-term advancement is hindered by the lack of systematic accumulation and transfer of skills. Without a unified mechanism for skill consolidation, agents frequently “reinvent the wheel”, rediscovering solutions in isolated contexts without leveraging prior strategies. To address this challenge, we introduce SkillNet, an open infrastructure for creating, evaluating, and organizing AI skills at scale. SkillNet structures skills within a unified ontology that supports creating skills from heterogeneous sources, establishing rich relational connections, and performing multi-dimensional evaluation across Safety, Completeness, Executability, Maintainability, and Cost-awareness. Our infrastructure integrates a repository of over 600,000 skills, an interactive platform, and a versatile Python toolkit. Experiments on ALFWorld, WebShop, and ScienceWorld show 40
Flexible perovskite light-emitting diodes (f-PeLEDs) hold great promise for next-generation optoelectronics, but their efficiency and operational half-lifetime (T 50) remain limited by poor film quality on deformable substrates. Herein, we develop a polymer interfacial permeation strategy by incorporating poly(2-ethyl-2-oxazoline) (PEOXA) into the hole transport layer (HTL). Through an antisolvent-free one-step spin-coating process, PEOXA spontaneously partitions between the HTL and perovskite emitting layer, forming a 3D interpenetrating network that simultaneously passivates defects and enhances mechanical flexibility. The carbonyl groups in the polymer chains act as Lewis bases, simultaneously improving luminescence efficiency and enhancing crystallinity. Meanwhile, the polymer incorporation reduces the Young's modulus, significantly boosting device flexibility. These synergistic effects enable green f-PeLEDs to achieve a record-high peak external quantum efficiency of 23.4% along with a power conversion efficiency of 26.7%, and demonstrate exceptional operational stability, with a measured T 95 lifetime of 3.2 h at an initial luminance of 1000 cd/m2. This work thereby establishes a versatile and robust strategy for the development of high-performance, flexible perovskite optoelectronics.
