Ionic skins (I-skins) integrated with high-performance electromagnetic interference (EMI) shielding, self-powering functionality, and sensitive pressure detection are critically needed for soft electronics, yet remains a considerable challenge. Herein, we report a facile diffusion-complexation strategy for fabricating self-oriented gradient MXene/polyelectrolyte hydrogels, which function effectively as self-powered I-skins with outstanding EMI shielding and pressure-sensing capabilities. A pivotal feature of our method is the concurrent coupling of two processes: osmotic pressure gradient- induced in-plane self-orientation of MXene nanosheets and diffusion-complexation reaction-induced spontaneous formation of a longitudinal charge gradient. This charge gradient endows the I-skins with a pressure-sensitive self-polarized potential, while the in-plane aligned MXene network imparts both exceptional EMI shielding performance and high-pressure sensitivity. Consequently, the resulting I-skins exhibit a high EMI shielding effectiveness of 48 dB, along with outstanding self-powered sensing capability that delivers a high sensitivity (3.067 mV kPa⁻¹), a broad sensing range (0.05–80 kPa), a fast response time (120–130 ms), and a low detection limit for pressure (0.05 kPa). This work offers a new and scalable strategy for integrating high sensitivity, energy autonomy, and superior EMI shielding into advanced I-skins for broader applications.

As heat waves intensify, helping crops survive sudden temperature spikes has become a practical goal for vegetable production. 

The commercialization of liquid lithium-ion batteries has revolutionized the consumer electronics industry. However, conventional lithium-ion batteries with graphite anodes and organic electrolytes are approaching their intrinsic performance limits and struggle to meet the growing demands for higher energy density, reliability, and safety in electric vehicles and large-scale energy storage. Solid-state batteries utilizing lithium or sodium metal anodes are considered promising next-generation energy storage solutions. Despite this potential, the formation of dendrites during charge–discharge cycling remains a critical challenge. Dendrite growth can initiate a destructive feedback loop of crack propagation and further dendrite intrusion, ultimately leading to battery failure and performance degradation. Previous studies have predominantly focused on single physical domains, such as electrochemical, stress, or thermal fields. However, such single-physics approach limits the understanding of dendrite evolution under realistic, coupled multiphysics conditions. This review first compares the morphological characteristics of dendrites in liquid and solid-state metal batteries. It then critically examines the key factors and predictive models of dendrite formation, initially from single-physics and subsequently from an integrated multiphysics perspective. Finally, strategies for mitigating dendrite growth via multiphysics field regulation are summarized. By establishing a comprehensive framework that integrates morphology evolution, multiphysics modeling, and suppression strategies, this work provides a foundational theoretical understanding for addressing dendrite formation in solid-state lithium and sodium metal batteries.

Fluorescence molecular imaging is an evolving technology for precision tumor diagnosis and therapy. By means of fluorescence probes, it enables the visualization of tumor-associated molecules, cells, or microenvironmental components, thereby facilitating the efficient identification and localization of lesions. With recent advances in the near-infrared II (NIR-II) window, targeted probes, microenvironment-responsive probes, multimodal imaging strategies, and theranostic integrated probes, fluorescence molecular imaging is shifting from a modality primarily used to “visualize lesions” toward a more comprehensive platform for tumor identification, therapy facilitation, and treatment-response monitoring. Meanwhile, artificial intelligence, particularly deep learning, can support image denoising, enhancement, resolution improvement, and three-dimensional localization, and may further assist in tumor recognition and grading, minimal residual disease assessment, and metastatic risk estimation. Overall, the integration of fluorescence imaging with AI is advancing oncology toward a more precise, intelligent, and integrated paradigm; however, key challenges remain, including probe biocompatibility and safety, clinical translation, and the lack of standardized imaging protocols.