Jaehyun Hur

Progress in sustainable hydrogen production requires electrocatalysts that accelerate the hydrogen evolution reaction (HER) with excellent catalytic activity and durability. Single-atom catalysts (SACs) have gained attention as superior HER electrocatalysts because of their almost 100% atomic utilization, distinct coordination surroundings, and tunable electronic structures. This review surveys advances in HER SACs with emphasis on atomic-level engineering and catalytic mechanisms. We first explore the fundamental principles and reaction mechanisms of the HER. The morphological engineering of numerous SAC architectures is discussed, along with their structural features and roles in boosting HER performance. The structural characteristics of SACs are examined to explain the roles of orbital interactions, charge redistribution, local structural asymmetry, and metal–support interactions in regulating hydrogen adsorption and HER kinetics. Finally, the catalytic performance of several SAC structures is discussed extensively, with a focus on HER activity, reaction kinetics, and prolonged stability. Density functional theory (DFT) is used to elucidate the mechanism underlying the outstanding catalytic performance of SACs in the HER. Moreover, key challenges associated with scalability, catalyst stability, and long-term operation at practical current densities are discussed, and future research directions are proposed. Finally, we discuss the rational development of next-generation SACs toward sustainable hydrogen production.

Neuromorphic devices are increasingly considered for edge‐intelligence systems operating in spaceborne, aerospace, nuclear, cryogenic, and other harsh environments, where local decision‐making is constrained by limited power, communication bandwidth, and latency. Reliable operation in these environments requires more than the preservation of basic electrical functionality because neuromorphic computation is physically encoded in material and interfacial state variables. Environmental stress can reshape these variables and alter analog modulation, plasticity, memory retention, threshold dynamics, and readout stability. This review discusses neuromorphic devices for extreme environments from the perspective of state‐variable reliability and stabilization. Radiation, temperature extremes, and vacuum or low‐pressure conditions are examined as major stressors that perturb defect generation, ionic and electronic transport, interfacial chemistry, switching pathways, and signal readout. Emerging space‐relevant factors, including microgravity‐assisted fabrication and mechanical loading, are also considered with respect to device formation, structural integrity, and long‐term reliability. Stabilization strategies are organized across material‐level design, environmental boundary control, and programming or readout compensation. By connecting device‐level degradation to neuromorphic metrics, array operation, soft‐error response, and inference accuracy, this review provides a framework for evaluating and designing reliable neuromorphic hardware under realistic environmental constraints.

Organic and oxide semiconductors differ fundamentally in their transport physics, interfacial chemistry, mechanical behavior, and processing requirements, yet these differences also make them functionally complementary. Integrating the two material classes, however, remains challenging because of their distinct surface chemistries, deposition conditions, thermal budgets, and interfacial stability. Beyond the question of whether such dissimilar materials can simply coexist, a more important issue is whether their integration can create functions unavailable to either material alone. This review demonstrates how carefully engineered organic‐oxide heterointerfaces can transform material dissimilarity into a functional advantage, enabling sensing, memory, nonlinear signal processing, and computation. We classify hybridization at the circuit, channel, and interface‐layer levels and connect each architecture to its underlying charge‐injection, transfer, trapping, recombination, tunneling, and fluctuation mechanisms. Lateral and vertically stacked circuits exploit the natural polarity matching between p‐type organic and n‐type oxide semiconductors, whereas directly coupled or interlayer‐mediated channels enable ambipolar, anti‐ambipolar, negative‐transconductance, Gaussian‐like, and sigmoidal responses. Hybrid contact, dielectric, passivation, and charge‐storage layers further regulate injection barriers, molecular organization, photogating, and carrier retention. By linking hybrid architecture and interface coupling to device transfer functions and target applications, this review presents organic‐oxide heterointerfaces as a design platform for thin‐film transistors and heterojunction devices.

Artificial synapses capable of analog memory and adaptive learning are key components for neuromorphic computing systems. To emulate biological synaptic functions effectively, artificial devices must exhibit gradual conductance modulation, linear responses to pulse stimuli, diverse forms of synaptic plasticity, and low‐power operation. Ion‐mediated electrochemical devices have recently emerged as promising candidates for such functionalities because ionic redistribution can continuously modify the internal electrochemical state of materials and naturally produce history‐dependent conductance changes. This review discusses recent advances in battery‐ion‐inspired synaptic devices that exploit electrochemical processes to implement artificial synaptic behavior. We first outline the fundamental electrochemical mechanisms underlying ion‐mediated state modulation, including ion insertion/accumulation, intercalation, migration, and diffusion. We then survey the materials landscape for battery‐inspired synaptic devices, covering electrolytes and active channel materials such as transition metal oxides, two‐dimensional materials, and organic mixed ionic‐electronic conductors. Emerging applications in neuromorphic computing and neuromorphic biosensing are revisited, where ionic dynamics enable the integration of sensing, memory, and computation. Key challenges related to device reliability, scalability, and environmental stability are discussed, together with future perspectives for advancing battery‐ion‐based neuromorphic technologies.

The development of high-performance anode materials remains a key challenge for advancing lithium-ion battery (LIB) technology. In this work, an oxy-bridged polymer cobalt phthalocyanine (Poly-CoPc) and its single-walled carbon nanotube-modified composite (Poly-CoPc/SWCNT) were successfully synthesized and systematically investigated as potential anode materials. The structural, chemical, and morphological properties of the materials were thoroughly characterized using Fourier-transform infrared spectroscopy (FT-IR), Raman spectroscopy, X-ray diffraction (XRD), Brunauer–Emmett–Teller (BET) surface area analysis, scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (SEM-EDX), and transmission electron microscopy (TEM) techniques. Morphological studies demonstrated a uniform dispersion of Poly-CoPc over the SWCNT network, leading to an interconnected conductive structure with enhanced surface area. Electrochemical performance was evaluated and compared with pristine Poly-CoPc. Compared to pristine Poly-CoPc, the Poly-CoPc/SWCNT composite exhibited significantly improved electrochemical behavior, including higher specific capacity, enhanced rate performance, and superior cycling stability. The composite delivered a high initial discharge capacity of 2016 mA g⁻¹ and maintained an excellent reversible capacity of 1047 mA g⁻¹ after 100 cycles at 0.1 A g⁻¹. Furthermore, it demonstrated outstanding rate capability, retaining a capacity of 855 mA g⁻¹ at 0.5 A g⁻¹. This enhanced performance is attributed to the synergistic effect between the redox-active Poly-CoPc and the highly conductive SWCNT network, which facilitates efficient electron transport, improves ion diffusion, and stabilizes the electrode structure during cycling. These results highlight that SWCNT-modified Poly-CoPc is a promising candidate for next-generation high-performance LIB anodes.

Intelligent systems operate in environments where multiple physical stimuli coexist and interact, yet most optoelectronic synaptic devices have primarily been limited to single‐modal operation. This review expands optoelectronic synapses beyond conventional optical‐electrical modulation toward a multisensory neuromorphic framework that incorporates thermal, mechanical, chemical, acoustic, and other physical stimuli. In multimodal optoelectronic synaptic systems, sensing, memory, and computation are coupled through conductance‐state evolution. External stimuli perturb internal variables such as photocarrier density, trap occupancy, ion distribution, interfacial polarization, adsorption‐induced charge, or phase state. The resulting changes in these internal states modulate conductance, producing volatile responses for transient sensory encoding or nonvolatile states for memory storage and weighted readout. We systematically review multimodal plasticity in optoelectronic synapses from both mechanism‐oriented and architecture‐based perspectives. Fundamental mechanisms are classified into five types: molecular structural transformation, filamentary switching, band structure engineering, field‐induced charge modulation, and carrier generation pathways. In this review, the reported multimodal optoelectronic neuromorphic devices are classified into five architectural types according to where transduction, memory, and computation occur, including single‐device integration, parallel multimodal processing, sensor‐coupled preprocessing, cross‐modal conversion, and stimulus‐dependent bidirectional plasticity. Representative applications in pattern recognition, adaptive learning, and in‐sensor/in‐memory computing are discussed as fundamental demonstrations toward realizing hardware‐level multisensory intelligence.

Strongly quantum‐confined CsPbI ₃ quantum dots (QDs) with an average size of approximately 5 nm are particularly well‐suited for achieving pure‐red emission that meets the stringent Rec.2020 gamut standard. The external quantum efficiency of small CsPbI ₃ QD‐based light‐emitting diodes (LEDs) has improved dramatically owing to the sustained efforts of the research community. However, their practical applications remain limited owing to their poor operational stability and low luminance. This review summarizes recent advances in high‐performance pure‐red LEDs based on CsPbI ₃ QDs and outlines potential directions for achieving red emission from strongly confined QDs with Commission Internationale de l'Éclairage coordinates of (0.708, 0.292). In addition, this review also highlights recent developments in CsPbI ₃ covering the deep‐red (635–650 nm) and near‐infrared‐edge spectral regions (650–700 nm) to provide better clarity and a more comprehensive understanding of the field. This article aims to serve as a valuable reference and stimulus for further interdisciplinary research in this rapidly evolving field.

Uncontrolled fires, from wildlands to industrial facilities, have become a pressing global threat, causing widespread ecological damage, loss of life, and economic disruption. The fundamental challenge is the lack of rapid and reliable detection at the ignition stage. Once flames spread, suppression becomes increasingly difficult and damage escalates. Conventional methods such as smoke detectors and thermal imaging fall short for wildfire and large-scale fire scenarios. Effective systems require accurate detection without false activation, stable performance under varied and harsh environments, and low power or zero-bias photodetection for continuous use in remote locations. Here, we report a wireless and flexible deep-ultraviolet (DUV) sensing platform that addresses these requirements in a single integrated unit. The platform combines a zinc tin oxide nanocomposite photodetector, flexible circuit integration, portable power, and Bluetooth communication. The sensor shows a selective solar-blind DUV response, stable operation under mechanical stress and extended cycling (92.5% retention after 100 bending cycles and 96.7% after 180 days), and energy-efficient performance compatible with autonomous deployment. Data-driven analysis of response curves allows for machine learning models that classify flame types and estimate distance, extending the system beyond binary fire detection by providing additional information on flame type and relative distance. This integrated approach provides a practical route to reliable fire monitoring, relevant to early-stage fire monitoring concepts for wildfire and industrial safety applications.