During lithium-ion battery thermal runaway, oxygen-depleted, highly concentrated flammable fumes are released, posing a significant challenge to safety control. Catalytic oxidation requires both sufficient O₂ and elevated temperatures for activation. However, thermal runaway off-gas is inherently oxygen-free, creating a fundamental paradox: direct air supply cools the gas below activation threshold, while pre-heating demands external energy input. To address this issue, this study proposes a chemically self-powered pre-activation strategy based on both O₂ and heat supply. By placing superoxides upstream of the catalyst and utilising their exothermic reaction with CO₂ in the fumes, the necessary O₂ and reaction heat were released during the initial stage of catalytic oxidation. Simultaneously, a blower is introduced to provide a continuous O₂ supply for subsequent sustained oxidation. This strategy achieves purification efficiencies exceeding 87% for both H₂ and CO throughout the thermal runaway event with zero open flame occurrence and outlet temperatures below 100°C, ensuring safe discharge. This study provides a viable technical pathway for the post-treatment of battery thermal runaway emissions, eliminating dependency on external heating or compressed oxygen supplies.

Self-standing hybrid solid polymer electrolytes (HSPEs) are herein fabricated through a solvent-free process, enabling energy-efficient and industry-compatible production. The HSPEs are based on a poly(vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP)/Pyr₁₄FSI ionic liquid (IL) system incorporating a small amount of divinyl sulfone (DVS) as a functional additive. This design delivers high ionic conductivity together with excellent mechanical integrity and thermal stability. The IL promotes efficient ion conduction and contributes to thermal stability, while incorporating DVS in little amounts (up to 5%) effectively stabilizes the solid–electrolyte interface (SEI) layer and mitigates degradation processes commonly observed in PVDF-HFP/Pyr₁₄FSI-based systems. This study presents the first demonstration of a PVDF-HFP/Pyr₁₄FSI-based electrolyte in which the incorporation of a small amount of DVS enables stable and reversible lithium plating/stripping over 1200 h of continuous cycling at room temperature, with a fixed areal capacity of 0.2 mAh cm^-2, in laboratory-scale solid-state lithium metal batteries (SSLMB). Furthermore, laboratory-scale SSLMB cells employing high-energy NMC811 cathodes exhibit excellent electrochemical performance, delivering specific capacities of 120 mAh g^-1 at C/20 and 100 mAh g^-1 at C/5 under ambient conditions, thereby advancing the development of high-voltage, intrinsically safe, and high-performance next-generation SSLMB technologies.

Perovskite solar cells have emerged as promising candidates for next-generation photovoltaics. However, optimizing their efficiency and long-term stability is complicated by the vast parameter space involving composition, processing, and device architecture, and so on. Conventional trial-and-error approaches are labor-intensive and often inefficient for navigating these complex multidimensional interactions. Consequently, the field is shifting toward high-throughput (HT) data-driven methodologies capable of accelerating the development cycle. This review highlights recent progress in HT computational and experimental strategies, along with their synergistic integration. Regarding computational efforts, we summarize the use of density functional theory and machine learning to identify single, double, and derivative perovskite candidates, as well as emerging techniques in literature data mining. We also discuss automated experimental platforms utilizing automatic processing and combinatorial characterization to optimize material compositions, device fabrication processes, and operational stability. Furthermore, we examine the integration of these domains through closed-loop workflows that combine computational prediction with automated experimentation to establish intelligent prediction, fabrication, and validation cycles. Finally, we provide an outlook on prevailing challenges regarding data acquisition, machine learning integration, HT platform infrastructure, and multi-modal data processing, alongside potential solutions for achieving fully autonomous material discovery and device optimization.

Lithium-sulfur batteries (LSBs) are considered promising candidates for high-energy-density storage systems. However, challenges such as the polysulfide shuttle effect and limited cycling stability hinder their commercialization. This study introduces a hierarchically structured interlayer, CNT @Co-N-CNF (CCNC), fabricated through a novel process that combines electrospinning with ZIF-67-derived carbonization. This interlayer uniquely integrates physical and chemical confinement mechanisms for lithium polysulfides (LiPSs) by leveraging a conductive carbon nanotube (CNT) network, nitrogen (N)-doping, and embedded cobalt species. Designed to enhance sulfur utilization and effectively suppress LiPS migration, the 50-CCNC interlayer demonstrates an initial discharge capacity of 1137.4 mAh g_s^-1, along with excellent cycling stability in coin cell configurations. Furthermore, its practical potential is validated in a pouch stack cell configuration, achieving an initial capacity of 1082.7 mAh g_s^-1and retaining 92.46% of its capacity after 40 cycles. These findings underscore the interlayer's effectiveness in addressing key challenges for large-scale LSB applications and represent a significant step toward the commercialization of high-energy-density LSB technology.

Interfacial instability in solid-state sodium storage is largely dictated by uneven ion flux and localized degradation, while its control through manufacturing parameters remains limited. Here, controlled densification is employed to regulate interfacial structure and sodium-ion behavior in nitrogen-doped bio-derived carbon electrodes. Increasing pressure leads to a transition from porous and discontinuous interfaces to compact and continuous pathways, which moderates ion flux and suppresses local Na⁺ accumulation. Electrochemical impedance measurements show a reduction in interfacial resistance from 320 to 140 Ω, accompanied by restrained resistance evolution during extended cycling. Structural and post-cycling analyses indicate that this stabilization is associated with more uniform ion redistribution and reduced defect formation at the interface. Nitrogen functionalities further contribute by tuning the interfacial electronic environment, supporting more stable ion transport. The optimized electrodes maintain capacity retention above 90% with consistent rate behavior. These observations reveal a direct link between densification, ion redistribution, and interfacial stability, indicating that ion transport can be regulated through manufacturing-controlled structural design. This work highlights a practical route for stabilizing solid-state interfaces through process-driven control of material architecture.

While mobile battery energy storage systems (MBESSs) are typically used to improve the stability of power systems, their ability to move also creates good opportunities for businesses to earn money through energy arbitrage. This profit depends heavily on decisions about timing and location, and is affected by uncertain conditions like fluctuating electricity prices and traffic. However, finding the best real-time control strategy that considers long-term profit and these uncertainties requires significant computing power. To tackle this issue, this paper presents a deep reinforcement learning framework for MBESSs designed to get the most profit from market arbitrage. Within this framework, we introduce the Knowledge-Assisted Deep Deterministic Policy Gradient (KA-DDPG) algorithm to learn the best policy more efficiently. The core novelty of KA-DDPG lies in its probabilistic hybrid action selection mechanism that unifies the agent's learned policy, offline expert criteria, and random exploration to manage the complex hybrid action space. Additionally, a two-phase guidance strategy is implemented to transition from offline-based to real-time-based criteria actions, ensuring both learning acceleration and policy robustness under computational constraints. Our rigorous statistical evaluations demonstrate that the proposed KA-DDPG approach leads to a 3%–7% improvement in average profits over the state-of-the-art Soft Actor-Critic baseline. Furthermore, it achieves exceptional policy stability, exhibiting a variance reduction of over 60% compared to standard DRL baselines and over 92% compared to the deterministic closed-loop MPC. The KA-DDPG algorithm also substantially speeds up the learning phase, validating its efficacy for real-time MBESS control under high uncertainty.