In recent years, supramolecular self-assembly has garnered increasing attention as a versatile strategy for materials design, offering novel opportunities in the development of functional biomaterials. Among these, natural herb extracts (NHEs) have emerged as promising building blocks due to their intrinsic ability to self-assemble through diverse noncovalent interactions. Compared with conventional polymer-based medical materials, NHE-derived assemblies possess distinctive advantages, including high drug-loading capacity, controlled-release behavior, permeability across biological barriers, and excellent biocompatibility. Notably, NHE self-assembled systems have demonstrated broad therapeutic potential, ranging from wound healing to cancer treatment. This review summarizes recent progress in the field of NHE-based supramolecular self-assembly. We first introduce the fundamental concepts of supramolecular self-assembly, highlighting the advantages and driving forces underlying NHE-derived assemblies. The major types of NHE molecules, their dominant interactions, and the resulting supramolecular architectures, particularly nanoparticles and hydrogels, are systematically reviewed, with emphasis on structure-function relationships and biomedical applications. In addition, advances in integrating NHE self-assembly with emerging technologies such as microneedles, photodynamic therapy, and photothermal therapy are discussed. Finally, we outline the current challenges and future research directions, aiming to provide insights into the rational design and translational potential of NHE-based supramolecular assemblies in biomedicine.
The ability to detect weak magnetic fields at femtotesla (fT) levels remains a critical obstacle for advancing fields like biomedicine and cross-medium communication. Existing solutions have faced fundamental limitations in operation environments, power efficiency, and sensitivity. Magnetoelectric (ME) sensors offer promising room-temperature alternatives but suffer from intrinsic material noise and extrinsic circuit limitations. Here, we propose a multi-parameter synergistic optimization strategy bridging material design and circuit integration, and establish a theoretical framework for laminated ME composites analysis. Integrated theoretical modeling and experimental validation identify material noise and amplifier noise as dominant noise sources. We go beyond the constraints of material noise and develop multi-longitudinal-transverse (MLT) mode laminates, featuring longitudinally magnetized magnetostrictive layers and transversely polarized piezoelectric fibers connected electrically in series. Compared to LT-mode (longitudinal-transverse) laminates, this design shifts the effective impedance to a reduced equivalent capacitance and increased DC (Direct Current) resistance regime, enabling direct tuning of dielectric loss noise. A single MLT laminates comprising seven LT mode units in a series configuration, achieved an equivalent magnetic noise of 3.84 pT/Hz1/2 at 1 Hz. Multiple MLT laminates of parallel array configurations resulted in a threefold reduction in equivalent magnetic noise compared to the series array configuration. Further, by implementing eight MLT units of parallel array configurations, we achieved a record equivalent magnetic noise of 927.8 fT/Hz1/2 at 1 Hz—surpassing noise performance of fluxgates and approaching SQUID sensitivity without cryogenic requirements.
The global challenge on energy crises has promoted the extensive research on energy storage, which positions the lithium-ion batteries (LIBs) as a core technology due to the exceptional performance. While conventional approaches for LIB research are limited by long development cycles, high resource demands, and heavy reliance on the expertise of researchers, data-driven methods powered by ML exhibit remarkable potentials on enabling the efficient and accurate predictions of key battery performance metrics, significantly reducing the time and economic costs on the R&D of batteries. Despite these advances, data scarcity remains a major obstacle to the widespread applications of these techniques due to the high costs and prolonged cycles associated with battery experiments. To investigate appropriate approaches for addressing these challenges, a comprehensive review on the data-driven methods for lithium battery research is provided, offering insights into mitigating the impact of data scarcity. To reach this goal, this review systematically examines the implementations of data-driven methods associated with the applications in the battery domain. Detailed studies have been conducted to investigate the root causes of data scarcity at various levels ranging from materials to devices and effective strategies to accommodate these issues. Finally, the review discusss future directions, emphasizing the need for collaborative data-sharing frameworks and adaptive ML models that balance domain expertise with computational innovation. By addressing data scarcity through these strategies, this work aims to accelerate the development of next-generation LIB while retaining the core insights of traditional methodologies.
CaTiO3:Tm (CTT) has unusual red-light charged near-infrared (NIR) persistent luminescence (PersL) property at ~ 800 nm. In this study, trap engineering was realized by co-doping CTT with 8 non-luminescent trivalent ions. Some basic laws in optimizing the NIR PersL of CTT were established based on the observation that rare earth dopants, including Sc/Y/La/Gd/Lu, at the A site of the ABO3 perovskite structure of CaTiO3 increased the PersL while the boron group dopants, including Al/Ga/In, at the B site led to decreased PersL. Y was found as an efficient co-dopant to realize enhanced NIR PersL in CaTiO3:Tm,Y (CTT-Y) with a ~60% enhancement vs. CTT. Thermal-stimulated luminescence (TSL) study indicate that Y is the most efficient element in generating abundant energy traps while slightly increased the trap depth from 320 to 326 K. Further, by choosing Pr as the 3rd dopant, visible-NIR double band fluorescence and PersL was realized in CaTiO3:Tm,Y,Pr (CTT-Y-Pr) at 613 and ~800 nm, which were ~5 and 2 times of those of CaTiO3:Tm,Pr (CTT-Pr), respectively. CTT-Y-Pr was applied to realize four-modal optical anti-counterfeiting imaging in vitro including visible fluorescence, visible PersL, invisible NIR fluorescence, and invisible NIR PersL. Moreover, visible fluorescence and NIR PersL imaging in vivo was applied to observe implanted tissue filler and its post-surgical tissue cleaning by using CTT-Y-Pr as an optical imaging contrast reagent, which was hard to retrieve once it was injected within body and formed unwanted tiny debris.
The leaky integrate-and-fire (LIF) neurons implemented in hardware have been proposed as a key approach for neuromorphic computing, offering a promising pathway to overcome the limitations of traditional von Neumann architectures. Among various candidates, ferroelectric-based neuromorphic devices (including antiferroelectric devices) offer a compact, energy-efficient, and highly scalable neuromorphic hardware, making them promising candidates for LIF neurons. This review systematically explains ferroelectric-based LIF neurons, covering the fundamental principles of neuronal operation, the implementation of neuronal functionalities, the key performance metrics, and strategies for performance optimization. Specifically, the implementation of neuronal functionalities is discussed focusing on the realization of leaky behavior by introducing depolarizing or inducing antiferroelectric phase to achieve volatility, since the neuronal integration and firing behaviors can be easily mimicked through the inherent cumulative polarization switching. Moreover, the key performance metrics, including hardware cost, energy consumption, and endurance of devices are identified to demonstrate the comprehensive advantages of ferroelectric LIF neurons. Additionally, the review also covers the applications of ferroelectric LIF neurons. Finally, this review summarizes challenges and prospects of ferroelectric-based artificial neurons for advanced neuromorphic computing systems. This review aims to provide theoretical guidance and practical insights to support further progress in neuromorphic computing systems based on ferroelectric materials.
Oral probiotic-based therapy has emerged as a promising solution with multifaceted benefits for inflammatory bowel disease (IBD) treatment. However, their widespread and clinical utility is severely limited by the poor viability of probiotics under harsh gastrointestinal conditions and elevated oxidative stress in the inflamed intestine. To address these challenges, a probiotic-based biohybrid (Lp@AU@GN) was bio-orthogonally fabricated by covalently anchoring the gold nanocluster-based artificial enzyme (AU) to the probiotic Lactobacillus plantarum (Lp), a strain screened out with the IBD-alleviating potential, followed by covalent encapsulation with the prebiotic β-glucan (GN). Upon oral administration to mice with ulcerative colitis, GN performed as a shield to physically protect Lp from gastrointestinal stress insults. After reaching the intestine, GN was metabolized by the gut microbiota, facilitating the exposure of Lp@AU and the concurrent production of short-chain fatty acids (SCFAs). The incorporated AU with superoxide dismutase- and catalase-like activities efficiently scavenged excessive reactive oxygen species in situ, neutralizing oxidative stress and simultaneously enhancing Lp survival to synergize with SCFAs to advance the therapeutic process. Consequently, therapeutic benefits of reduced inflammation, restored intestinal barrier, and rebalanced gut microbiota were achieved. Furthermore, the therapeutic utility was extended to Crohn's disease, establishing the broad-spectrum effectiveness of Lp@AU@GN for addressing both major forms of IBD. Beyond IBD, our developed modular engineering strategy holds the great potential to be adapted to fabricate diverse biohybrids for other gastrointestinal or metabolic disorders’ treatments by tuning the probiotic strain, antioxidant moiety, or prebiotic polymer.
