Shenzhen Institute of Building Research (China)

ABSTRACT Conversion‐type electrodes with extended Li + storage and multi‐electron transfer capabilities hold great promise for meeting the growing demand for high energy density. However, most of these materials suffer from inherently poor rate performance and limited capacity reversibility, primarily due to the low diffusivity of metal cations and inevitable structural changes during (de)lithiation. Herein, we report the high first‐cycle capacity reversibility of FeS 2 under ultrahigh charging rates. The kinetic properties of the charging process are revealed through electrochemical measurements, and the origin of the outstanding rate performance is further investigated using advanced spectroscopic characterizations. In particular, the exceptionally high apparent diffusion coefficient observed in the electrode reflects the excellent mobility of Fe 2+ cations, which directly accounts for the remarkable rate capability. Further investigation reveals that the S 2 – anion framework of the anti‐fluorite structure is well preserved during charging, providing stable channels and sites for Fe 2+ insertion. Moreover, the high‐spin state of Fe 2+ reduces the energy barrier for its migration into the tetrahedral sites formed by S 2 – anions, facilitating rapid and stable ion transport. This study elucidates the origin of the kinetic performance from a mechanistic perspective and provides guidance for the development of conversion‐type electrodes in high‐power energy storage devices.

This paper presents Reed-Solomon coded single-stranded representation learning (RSRL), a novel end-to-end model for learning representations for lossless DNA data storage. In contrast to existing learning-based methods, RSRL is inspired by both error-correction codec and structural biology. Specifically, RSRL first learns the representations for the subsequent storage from the binary data transformed by the Reed-Solomon codec (RS code). Then, the representations are masked by an RS-code-informed mask to focus on correcting the burst errors occurring in the learning process. The synergy of RS masks and graph attention enables active error localization, breaking through the limitations of traditional passive error correction. With the decoded representations with error corrections, a novel biologically stabilized loss is formulated to regularize the data representations to possess stable single-stranded structures. By incorporating these novel strategies, RSRL can learn highly durable, dense, and lossless representations for subsequent storage tasks in DNA sequences. The proposed RSRL has been compared with a number of baselines in real-world tasks of multi-type data storage. The experimental results obtained demonstrate that RSRL can store diverse types of data with much higher information density and durability, but much lower error rates.

Al/Mg bimetallic composites have drawn considerable attention for their promising lightweight applications in sectors such as the aerospace and automotive industries. In these systems, the interfacial behavior critically governs the overall performance and reliability. In this research, the molecular dynamics (MD) simulation was employed to systematically study the effects of pouring temperatures (923 K, 973 K, and 1023 K) and preheating temperatures (373 K, 473 K, and 573 K) on the interfacial diffusion behavior and fracture mechanism of the polycrystalline Al/Mg bimetallic system. The results indicate that the influencing rule of pouring temperatures and preheating temperatures on the interfacial diffusion behavior is consistent. Specifically, the diffusion coefficient of Mg atoms is higher than that of Al atoms, while the diffusion distance of Al atoms is significantly greater than that of Mg atoms. As the temperature increases, the thickness of the interfacial transition layer correspondingly rises. However, the effects of these two parameters on tensile fracture behavior demonstrate notable discrepancies. Specifically, the fracture mode evolves with pouring temperature, transitioning from being mediated solely by dislocations to being co-mediated by twins and dislocations. In contrast, the fracture mechanism remains solely dislocation-controlled, regardless of the preheating temperature. In addition, all the models fractured at the interface between the diffusion layer and the Mg matrix. The optimal tensile strength of 1.850 GPa was achieved at a pouring temperature of 923 K and a preheating temperature of 473 K, representing an improvement of approximately 52% compared to the lowest value recorded in the study. This research offers significant theoretical insights for the rational optimization of preparation parameters and an in-depth understanding of fracture mechanisms in Al/Mg bimetallic systems.

The widespread deployment of automated guided vehicles (AGVs) in dynamic complex environments poses a critical challenge in achieving efficient avoidance and energy optimization. This paper proposes an observation-enhanced reinforcement learning with reward constraints algorithm (ORLRCA) to enhance AGV safety and energy efficiency obstacle avoiance in complex scenarios. In our work, firstly, a multimodal perceptual attention mechanism is introduced to dynamically capture obstacle motion patterns and environmental semantic features, thereby enhancing scene perception capabilities. Secondly, a multi-scale prior reward-constrained function is designed to jointly optimize safety distance, path smoothness, and energy consumption metrics, effectively addressing the suboptimal convergence caused by conflicting objectives in RL strategies. Finally, leveraging the actor-critic network architecture of proximal policy optimization (PPO), we achieve end-to-end optimization of robust obstacle avoidance policy generation through synergistic integration of attention-enhanced state representations and multi-scale reward signals. To validate efficacy, a high-fidelity simulation environment is developed for comparative experiments. Results demonstrate that the proposed algorithm exhibits superior performance in obstacle avoidance success rate and energy efficiency compared to baseline RL methods, establishing a theoretical foundation for secure deployment and energy-efficient management of AGVs in practical industrial scenarios.

The magnetic proximity effect (MPE) allows nonmagnetic materials to acquire magnetic properties via interfacial exchange, providing a route to manipulate spin and valley degrees of freedom in 2D heterostructures. However, MPE is typically limited by weak interfacial coupling, constraining the control of valley polarization and excitonic states in high-performance valleytronic devices. Here, we demonstrate that the MPE in the Fe3GaTe2/WS2 heterostructure is effectively tuned as strain from a gold grating substrate enhances the interlayer coupling. The enhanced interlayer coupling promotes spin-dependent charge transfer in the MPE process, driving a pronounced longitudinal expansion of the WS2 exciton valley polarization hysteresis loop and achieving a substantial modulation amplitude within the ±1 T magnetic field range. Simultaneously, the strain-enhanced MPE induces a pronouncedly enhanced Zeeman splitting, driving a giant Landé g-factor of WS2 excitons up to -30. First-principles calculations attribute these effects to strain-induced band-structure reconfiguration and enhanced interlayer charge transfer. Our findings establish strain engineering as a powerful and precise approach to control the MPE, opening new avenues for advanced spin-valleytronic applications and next-generation low-dimensional quantum devices.