The clinical management of hemorrhagic patients remains a significant challenge, as current therapeutic approaches lack real‐time diagnostic capability and spatiotemporal precision. Conventional materials (e.g., hydrogels) require surgical intervention, while molecular targeting strategies are limited by slow kinetics and uncontrollable thrombotic risks. To address these limitations, we developed a smart nanomaterial based on a NiFe‐layered double hydroxide decorated with both Ag single atoms and nanoparticles (AgSA + NP/NiFe‐LDH). This platform can bind plasminogen activators, achieving rapid hemostasis with approximately 30% reduced fibrinolysis in vitro and approximately 50% reduced blood loss in vivo in models of intracranial and abdominal hemorrhage. Under pulsed near‐infrared (NIR) irradiation, the dominant photoelectric effect enhances hemostatic efficacy, reducing blood loss by about 50% compared to the unexposed part; under sustained NIR exposure, strong photothermal effects inactivate clot‐enriched thrombin and alter fibrin structure, thereby promoting controlled thrombolysis. This dual functionality is consistently demonstrated in arterial models and in cryopreserved human blood samples from healthy individuals and hemorrhagic patients. Moreover, the material enables real‐time visualization of active bleeding (as low as 3 μL) via photoacoustic imaging, as validated in an intracranial hemorrhage model. Our work demonstrates a proof‐of‐concept NIR‐based strategy for externally modulated management with potential applications in dynamic vascular emergencies.
Reliable organic ferroelectrics for memory applications require high endurance under repeated electrical switching. Here we demonstrate robust fatigue resistance in highly crystalline 2‐methylbenzimidazole (MBI) films grown by low‐temperature deposition followed by restrained crystallization (LDRC) on interdigital electrodes. Switching kinetics analyzed using the Kolmogorov–Avrami–Ishibashi (KAI) model reveals characteristic millisecond switching times and quasi‐one‐dimensional domain growth associated with proton transfer along hydrogen‐bond chains. Guided by these kinetics, we examined the switching endurance under a stringent protocol that ensures complete polarization reversal. After ~108 continuous cycles (~2 weeks), the remanent polarization shows a small initial wake‐up behavior (~10% increase within the first 104 cycles), followed by stabilization at values close to the initial polarization, indicating robust ferroelectric endurance under prolonged electrical cycling. The combination of LDRC‐enabled high crystallinity and localized proton‐transfer switching, which introduces minimal structural perturbation during polarization reversal, may have enabled this fatigue tolerance and highlight MBI as a simple, fluorine‐free platform for durable organic ferroelectric devices.
Multivalued logic (MVL), which encodes information beyond binary states, offers a transformative approach to enhancing computational efficiency in the post‐Moore era. Conventional MVL implementations typically rely on lateral heterostructures that exhibit negative differential transconductance (NDT) or negative differential resistance (NDR) behavior. However, such designs are often hindered by fabrication complexity and sensitivity to voltage thresholds. Here, we present a simple and effective strategy for MVL devices: a vertical heterojunction architecture with a self‐assembled monolayer (SAM)‐modified interface. The SAM not only passivates surface defects but also spatially confines hole carriers. This yields a distinctive V‐shaped transfer characteristic with suppressed hole conduction, a feature critical for achieving stable intermediate logic states without the need for intricate multithreshold engineering. We further demonstrate a ternary inverter with three distinct logic states, whose intermediate state range is dynamically tunable via supply voltage in the subthreshold regime. Our work provides a promising pathway toward energy‐efficient MVL systems for applications in artificial intelligence and edge computing.
Thin‐film transistors based on amorphous oxide semiconductors demand thinner channels for next‐generation display and memory applications. However, channel‐thickness scaling causes a positive threshold‐voltage shift and degrades field‐effect mobility. Here, we report film densification in sputter‐deposited, hydrogen‐free amorphous indium–gallium–zinc oxide as the channel thickness decreases, with the film density exceeding that of single‐crystal InGaZnO4. Ab initio molecular dynamics simulations reveal a change in polyhedral connectivity from corner‐sharing to edge‐sharing. Fourier‐transform infrared spectroscopy and chemical etch‐rate measurements support this structural change, showing a shift of the M─O vibrational band peak to lower wavenumber and an increased etch rate. We propose that the resulting reduction in the M─O─M bond angle induces strain. In the proposed model, this strained local structure can enhance the formation of interstitial oxygen and lead to the formation of an oxygen Frenkel defect pair. This defect pair deepens the donor level and broadens the tail states below the conduction‐band minimum, thereby degrading thin‐film transistor performance. The degradation is suppressed in the In‐rich composition. This result may be understood in terms of the stabilization of interstitial oxygen as an electrically inactive peroxide species identified by the experimental results.
Group III–V semiconductors are suitable for solar‐driven hydrogen production due to precise lattice matching (for tandem device combinations), high carrier mobilities (>103 cm2 V–1 s–1), tunable bandgaps (1.1–2.5 eV), strong absorption coefficients (>105 cm–1), and a large range of doping possibilities. GaInP2, GaAs, and InP have a theoretical solar‐to‐hydrogen efficiency of more than 20%. Still, their high susceptibility to chemical and photochemical corrosion during prolonged photoelectrochemical water splitting (PEC‐WS) is a major industrial hurdle. This review comprehensively covers critical strategies to enhance the photostability of group III–V semiconductors for industrial‐scale PEC‐WS, emphasizing artificial intelligence (AI) and machine learning (ML) support. We discuss three main intrinsic and extrinsic strategies that include (i) AI‐driven crystal structure modification and faceted growth control, (ii) physical and chemical surface passivation coupled with machine learning for passivation material screening strategies, and (iii) combining III–V semiconductors with appropriate band‐aligned materials via preferable heterojunctions, utilizing algorithms to select optimal compositions. We examine the underlying mechanisms of each strategy and conclude with perspectives on achieving optimal materials for PEC‐WS using advanced AI and ML screening tools. Implementing these effective modification strategies, the wide commercial potential of III–V photoelectrodes for PEC‐WS technology can be unlocked.
