High-quality-factor (Q-factor) mid-infrared (MIR) light sources with spectral tunability and minimal angular dependence are highly desirable for practical sensing applications. In this work, a continuously tunable, low-angular dispersion, and narrowband thermal emitter is realized by incorporating a Ge2Sb2Te5 (GST) layer into the grating-assisted distributed Bragg reflector (DBR)-like structure. The multilayer back reflector is designed to achieve high reflectance over a broad DBR-like stopband, while simultaneously avoiding direct contact with the metallic back reflector, as in conventional metal-dielectric-metal (MDM) structures. Upon illumination with transverse-magnetic (TM)-polarized light, the hybrid structure is found to support a hybridized localized surface plasmon (h-LSP) with a Q-factor of 30.64 in simulations (in contrast to the LSP mode in MDM structures with a Q-factor of 6.17) due to the reduction of ohmic loss. Meanwhile, the h-LSP mode shows a similar geometric dependence on grating width and exhibits a negligible wavelength variation under varying oblique incidence angles. Samples with three grating widths of 1.4 μm, 1.7 μm, and 2.0 μm were fabricated to support the h-LSP mode at distinct wavelengths. Upon thermal annealing, both the measured reflection and emission spectra of these samples exhibit a continuous redshift, with an average shift of 140 nm corresponding to an approximately 40% crystalline fraction of GST. Meanwhile, the Q-factors of the emission peaks remain above 8.70 for w = 1.4 μm, 12.72 for w = 1.7 μm, and 15.68 for w = 2.0 μm. Such a tunable, narrowband, and small angular-dependent MIR light source is promising for enhancing the accuracy in the discrimination of molecular fingerprints.

Surface-enhanced infrared absorption spectroscopy is a powerful tool for label-free molecular fingerprinting. However, its quantitative accuracy and sensitivity are often compromised by the inherent roughness of metallic substrates, which leads to heterogeneous probe adsorption and random molecular orientation. Herein, we propose a dual-interface engineering strategy to construct a robust biosensing platform for the highly sensitive detection of microRNA-21. A metal-insulator-metal plasmonic sensor is functionalized with monolayer graphene, serving as an atomically flat buffer layer to shield the underlying gold grain roughness and facilitate uniform bio-functionalization via non-covalent π-π stacking interactions. Furthermore, rigid tetrahedral DNA nanostructures are immobilized on the graphene surface as upright scaffolds, effectively regulating the probe spacing and preventing the "lying-down" orientation common to single-stranded DNA. This synergistic design not only maximizes the overlap between the target biomolecules and the plasmonic near-field hotspots but also significantly improves the signal-to-noise ratio and sensor reproducibility. Consequently, the sensor demonstrates a superior sensing capability detectable down to 1pM for miRNA-21, representing a 50-fold improvement over the bare metal-insulator-metal (MIM) counterpart. This work highlights the critical role of interfacial architecture in plasmonic sensing and provides a promising strategy for future bioanalytical applications.

Holography plays a vital role in optics, while conventional methods suffer from complex devices and bulky systems. Metasurfaces, versatile platforms for manipulating light at subwavelength scales, have enabled the realization of compact holographic devices. However, most existing metasurface holograms are limited to scalar fields, modulating only intensity while neglecting polarization information. While vectorial holography has attracted significant attention, particularly utilizing metal–insulator-metal (MIM) structures, achieving high-performance vectorial manipulation at visible wavelengths remains challenging due to high ohmic losses. Here, we demonstrate visible-wavelength vectorial holography using an MIM metasurface. Our approach combines an improved vectorial Gerchberg-Saxton (GS) algorithm with a wave-decomposition technique to design subwavelength meta-atoms. We perform full-wave simulations to demonstrate that it can generate target vectorial holographic images under illumination by circularly polarized light at 633 nm, providing an implementation case for vectorial light-field manipulation in the visible wavelength and demonstrating the potential for realizing advanced vectorial holographic displays in the visible wavelength using standard MIM structures.

Quantum tunneling offers a fascinating paradigm for orchestrating spatial charge transport in artificial photosynthesis. However, precisely manipulating electron tunneling across well-defined heterointerfaces remains a formidable challenge, with conventional designs largely confined to classical Semiconductor–Insulator–Metal (S–I–M) architectures. Herein, we report a conceptual endeavor by fundamentally departing from the traditional S–I–M model, constructing a unique and novel semiconductor–insulator–ligand/metal tunneling platform. Specifically, an ultrathin insulating poly(sodium 4-styrenesulfonate) (PSS) layer is engineered onto a transition metal chalcogenide (TMC, e.g., CdS) substrate. Subsequently, poly(diallyldimethylammonium chloride) (PDDA)-capped metal nanocrystals (M@PDDA, M = Au, Pd) are precisely anchored via electrostatic self-assembly, yielding well-defined TMC@PSS/M@PDDA heterostructures. Distinct from conventional systems, the PDDA ligands synergistically couple with the metal core to form an integrated, highly potent electron capture center driven by the Schottky-junction effect. This unique synergistic driving force triggers non-classical, directional electron tunneling from the photoexcited TMC substrate directly through the insulating PSS barrier. Benefiting from this advantageous quantum tunneling, TMC@PSS/M@PDDA heterostructures demonstrate significantly enhanced and multifarious visible-light-driven photoredox activities including selective organic transformations and H2O2 production. This work establishes an elegant conceptual paradigm for decoding and customizing quantum tunneling pathways, offering profound fundamental insights into advanced solar energy conversion.

Previous femtosecond pump-terahertz probe experiments revealed that the intrinsically coupled insulator-to-metal (IMT) and structural phase (SPT) transitions in niobium dioxide (NbO2) occur at two distinct fluence thresholds, exhibiting a novel dual-threshold feature. Here, we demonstrate that they are induced by two distinct types of electron-phonon interactions, different from the earlier interpretation of purely electronic IMT at the first fluence threshold (Fth1) and photothermal SPT at the second (Fth2). At Fth1, the photoexcited holes are self-trapped at specific niobium sites, assisted by thermal phonons. Such polaronic distortions drive the IMT into a bad metallic phase, beyond purely electronic effects. At Fth2, besides the IMT that occurs first, coherent phonons induce SPT from the zigzag Nb chains to linear isometric configurations, which enhances the metallicity via a nonthermal pathway rather than thermal effects. Our study provides theoretical insights into potential applications of NbO2 for multilevel and energy-efficient storage.

Accurate knowledgeof the electrical and thermal conductivitiesand structural properties of hydrogen–helium mixtures underthermodynamic conditions within and beyond the immiscibility rangeis very important to predict the thermal evolution and internal structureof gas giant planets like Jupiter and Saturn. Here, we propose a novelmethod to determine the immiscibility boundary accurately withoutthe need for free energy calculations, while providing consistentinsights into structural and transport properties of mixtures. Weshow with direct large-scale ab initio simulationsthat the insulator–metal transition (IMT) of the hydrogen subsystemis strongly affected by an admixture with a small fraction of heliumand occurs at temperatures significantly higher than those of purehydrogen. At pressures below 150 GPa, the IMT boundary is not relatedanymore to the H2 subsystem dissociation, the system remainsinsulating even after the full dissociation of H2 moleculesand its transition to an atomic H–He mixture. The offset ofthe IMT in the H–He mixture relative to the dissociation regionin the hydrogen subsystem and the significant reduction of staticelectrical and thermal conductivity by a factor between two and afew thousand relative to pure hydrogen found in mixtures have consequencesfor Jupiter and Saturn’s thermal evolution, internal structure,and dynamo action, affecting a large fraction of the interior of bothplanets.

The development of compact, high-performance plasmonic filters capable of multimode operation remains a critical challenge in integrated photonic sensing. In this work, a multimode bandpass plasmonic filter based on dual-spiral resonators is proposed and numerically investigated. The resonator is comprised of left- and right-handed variants and coupled to metal-insulator-metal (MIM) waveguides. Leveraging the self-similar geometry of spiral resonators, the design supports four spectrally distinct transmission peaks within the 600-1200nm near-infrared window, with high transmittance (up to 76.4%) and narrow full width at half-maximum (as low as 15.2nm). Using two-dimensional finite-difference time-domain (FDTD) simulations with a Drude-Lorentz model for silver, it is demonstrated that the spiral architecture enables independent spectral tuning of resonant modes through geometric parameters such as the winding angle and coupling gap. When employed as a refractive index (RI) sensor, the device exhibits sensitivities ranging from 801 to 896nm/RIU across its modes, with figure-of-merit (FoM) values ranging from 35.6 to 58.9RIU-1 across its modes. The compact footprint (520×540nm2) and robust multimode response make this platform highly suitable for label-free, multi-analyte biosensing and on-chip spectral multiplexing, offering a significant advance over conventional ring- or stub-based plasmonic filters.

All 2D van der Waals junction-FET biosensor with external capacitive sensing for rapid, label-free detection of liver cancer biomarkers.

This paper presents a metasurface with dual-band extrinsic chirality based on a metal-insulator-metal (MIM) structure. The metasurface utilizes mirror-symmetric split-ring resonators as fundamental building blocks. Under oblique microwave incidence, the in-plane mirror symmetry is effectively broken, inducing chiral electromagnetic resonance modes at two distinct frequencies. The selective coupling and suppression of left- and right-handed circularly polarized waves by the meta-atoms generate strong circular dichroism with opposite signs at the two frequencies. Concurrently, the different sensitivities of the resonant modes to incident angle variations result in mirror-symmetric frequency shifts for the two cross-polarization components within the bands, enabling flexible angular dispersion control. Under TM-polarized wave incidence, varying the incident and azimuth angles can excite significant phase dispersion, leading to rich polarization state evolution over the entire Poincaré sphere across the entire operating frequency band. By introducing the "incident angle" as a control dimension, this work provides a new, to the best of our knowledge, approach for multi-band, reconfigurable chiral responses and dynamic polarization manipulation, holding significant promise for applications in electromagnetic communication, polarization imaging, and related fields.

High-resolution optical sensors are vital for detecting subtle refractive index variations associated with biomolecular interactions, such as changes in carcinoembryonic antigen (CEA) concentration. This work presents a compact metal-insulator-metal (MIM) plasmonic sensor with two concentric ring resonators, numerically investigated using finite-difference time-domain simulations. Structural parameters are optimized to maximize field confinement and resonant mode coupling, yielding a sensitivity of 1035nm/RIU, a resonance linewidth of 3.75nm, and a figure of merit of 276 RIU⁻¹. The sensor demonstrates precise spectral shifts corresponding to minute refractive index changes linked to CEA, while the influence of surface-bound layers and fabrication tolerances is also assessed to ensure robustness under practical conditions. These results highlight the sensor's potential as a high-resolution, label-free platform for monitoring CEA in biological samples.

In this paper, a high-sensitivity humidity sensor structure based on a metal-insulator-metal (MIM) metasurface is proposed through numerical simulation. The top layer of the structure consists of a periodic array formed by two asymmetric grating units. By combining Fabry-Pérot (F-P) cavity resonance with bound states in the continuum (BIC) modes, and utilizing the swelling characteristics of the intermediate PVA hydrogel layer under different humidity conditions, sensitive humidity monitoring is achieved. Numerical simulation results show that under high relative humidity (>80%), the sensor achieves a peak sensitivity of 12.526nm/%RH, enabling accurate detection of humidity variations. Furthermore, the distribution of structural colors in the CIE 1931 chromaticity diagram under different humidity conditions covers the full color gamut, indicating that the relative humidity range can be assessed through color comparison. This study provides a feasible theoretical scheme for the design and application of high-sensitivity humidity sensors.

Bilirubin is an important biomarker, where a small unbound fraction dissociated from albumin can cross the blood–brain barrier and induce neurotoxicity, such as kernicterus, at low nanomolar levels. Accurate detection of this low-level fraction remains challenging. Surface-enhanced Raman spectroscopy (SERS) enables label-free molecular detection; however, variations in the local refractive index (RI) at plasmonic hotspots can detune the resonance from the excitation wavelength, leading to signal fluctuations and limited quantitative reliability. Here, we present a multi-resonant nanolaminate SERS substrate designed to achieve RI-insensitive and robust signal enhancement. The vertically stacked metal–insulator–metal architecture provides broadband spectral overlap with both excitation and Raman scattering under dielectric loading, maintaining consistent enhancement across varying RI conditions. We demonstrate label-free bilirubin detection with a highly linear response over 10−9 to 10−4 M, achieving an R2 value of 0.99. Compared with previously reported bilirubin SERS substrates relying mainly on single-resonant plasmonic enhancement, this RI-insensitive design offers improved quantitative reliability under dielectric environmental changes. These results highlight the importance of RI-insensitive SERS design for reliable quantification and provide a general strategy for robust SERS-based biosensing.

With the growing need for high-power density, high-efficiency power electronics, wide band gap (WBG) semiconductors, such as silicon carbide (SiC) and gallium nitride (GaN), have been widely used in recent years. With high switching speed, stray inductance induced by packaging would cause voltage overshooting and oscillation during the switching transient, which should be mitigated at all costs. In this paper, a power module design based on a multilayer insulated metal substrate (MIMS) structure was proposed to effectively address the stray inductance concern based on the mutual-inductance cancelling effect. Fabrication process flow with high feasibility was also designed. Electrical and thermal simulations were conducted based on a power module with a nominal rating of 1200 V and 500 A. Compared to the planar module, the proposed design possessed much lower stray inductance (3.47 nH vs. 14.85 nH). In the transient thermal simulation, the proposed module exhibited a time constant 141.7% higher than that of the hybrid module with a ceramic substrate on the bottom but MIMS on the top, making it suitable for applications with high-constant power output requirements.

The development of novel information-functional devices based on emergent physical phenomena is crucial for integrated circuit technology in the post-Moore era. Two-dimensional magnetic materials present an ideal platform for spintronic devices; however, regulating their room temperature magnetism poses significant challenges. Traditional methods like ionic liquid gating and strain control face issues such as poor stability and complex processes, complicating compatibility with standard silicon technology. Here, we demonstrate a straightforward and robust approach for dielectric layer-engineered room temperature ferromagnetism in 2D metallic magnets by leveraging metal–insulator–semiconductor (MIS) structures. Using surface-oxidized Fe3GeTe2 as a model system, we systematically investigate how SiOx dielectric layer thickness (50–300 nm) modulates magnetic properties. Thin dielectric layers significantly enhance room temperature ferromagnetism through boosted interfacial charge transfer, whereas thick layers maintain the material near its intrinsic state due to dielectric screening effects. Furthermore, reversible optical modulation of magnetism is achieved under ultraviolet illumination, with photoresponse capability diminishing as dielectric thickness increases. This work establishes a scalable, silicon-compatible strategy for controlling 2D magnetism and provides critical insights for developing optically tunable spintronic devices and non-volatile memory applications.

Single Molecule Magnets (SMMs) are promising for molecular spintronics owing to their quantum properties; however, reproducible integration into devices has remained challenging due to limitations of break-junction and gold-based systems. This study introduces a scalable NiFe/AlOx/p-Si Metal Insulator Semiconductor (MIS) platform that enables sidewall integration of lipoic acid-functionalized Mn₆ SMMs in a configuration consistent with sulfur-based molecular anchoring at exposed junction edges. Despite fabrication-related variability in pristine MIS junctions, molecular integration produced more convergent reproducible tunneling characteristics across multiple devices, as confirmed by standard deviation analysis. Importantly, charge transport in this architecture remains fundamentally governed by tunneling across the MIS barrier. Molecular integration does not replace intrinsic MIS transport but is associated with modification of interfacial electrostatics, electrode work function, and effective tunneling barrier properties, resulting in improved reproducibility of measured tunneling characteristics across devices. Kelvin Probe Force Microscopy (KPFM) revealed an approximately 0.4V increase in NiFe surface potential following SMM attachment, providing electrode-level evidence of interfacial electronic modification following molecular treatment. Conceptual modeling suggests two cooperative mechanisms: (i) molecule-induced modification of tunneling barrier electrostatics, quantitatively supported by Brinkman model analysis, and (ii) charge redistribution at the NiFe/SMM interface, which modifies electrode work function and electrostatic boundary conditions of the MIS junction. While direct spectroscopic identification of specific molecular orbitals such as the Lowest Unoccupied Molecular Orbital (LUMO) was not performed, the observed conductance enhancement, tunneling barrier modification, and surface potential shift strongly support molecule-induced modulation of the MIS tunneling barrier through interfacial electronic coupling and associated electrostatic barrier modulation. The present conclusions are limited to the specific case of lipoic acid-functionalized Mn₆ SMMs and should be interpreted as a system-specific demonstration rather than a universal statement for all molecularly modified MIS junctions.

Vanadium dioxide (VO2) films have attracted extensive attention for their pronounced metal–insulator transition (MIT) and multifunctional responses, holding great promise for smart windows, infrared stealth, memristive devices, and advanced sensors. However, conventional approaches for tuning the transition temperature, such as elemental doping or heterostructure engineering, often suffer from complicated processing, impurity phases, and poor device uniformity. Here, we use a dopant-free, high-vacuum annealing (9 × 10−4 Pa, ≈9 × 10−6 mbar) strategy to regulate the intrinsic structural evolution of VO2 films via oxygen-vacancy engineering and to clarify its influence on electrical switching contrast and infrared emissivity modulation. As the annealing temperature increases under low oxygen partial pressure, oxygen vacancies gradually accumulate, converting V4+ to V3+ and driving the films through three distinct structural stages: low-temperature lattice expansion with preserved M1 framework, critical structural collapse at 550 °C, and high-temperature defect rearrangement with local recrystallization. Consequently, the electrical MIT temperature continuously decreases, but the switching ratio collapses at the critical point and only partially recovers after high-temperature reorganization, while the infrared emissivity response transitions from abrupt, phase-transition-dominated switching to a continuous, tunable modulation at elevated temperatures. Notably, the infrared response begins continuous tuning earlier (≈450 °C) than the collapse of electrical MIT, reflecting the different sensitivities of optical and electronic responses to local lattice defects. These results reveal the coupling among oxygen-vacancy evolution, structural stability, electrical contrast, and infrared modulation in compositionally simple VO2 films. Compared with conventional doping, this high-vacuum annealing strategy avoids impurity phases, preserves compositional simplicity, and provides a scalable defect-engineering route to design VO2-based devices with reconfigurable electrical and infrared response modes.

The advancement of compact and wearable optical systems is critically limited by the lack of high-performance, integrable solutions for managing stray light, particularly in the mid-infrared regime. Traditional approaches relying on bulky baffles or cryogenic cooling hinder miniaturization and flexibility. Herein, we report a large-area, flexible metamaterial film based on a Ti/Al2O3/Fe3O4/Ti heterostructure that functions as an ultra-broadband infrared extinction layer. By introducing a lossy Fe3O4 interlayer, our design achieves an exceptional average absorptivity of 97.1% across the 3-5µm atmospheric window, with a total thickness below 1µm. This "loss engineering" strategy effectively broadens the absorption bandwidth and smoothens the spectral response compared to conventional metal-insulator-metal(MIM) absorbers. The film is fabricated on a flexible polyimide substrate via scalable lithography, demonstrating remarkable mechanical robustness. More importantly, we integrate it into a practical optical system as a cylindrical baffle, where it suppresses stray light intensity to a mere 0.6% after three reflections, significantly enhancing imaging contrast. This work presents not merely a high-performance absorber, but a scalable, flexible material platform that paves the way for next generation miniaturized and flexible optical devices.

Abstract The design of high-performance, stable transistors based on two-dimensional materials such as molybdenum disulfide ($MoS_2$) demands a thorough understanding of how disorder, electrostatic gating, and thermal effects shape electron transport. To move beyond conventional simulation approaches, this study introduces a novel framework that combines quantum chaos theory with machine learning to model and predict the behavior of $MoS_2$ field-effect transistors. Using the statistical distribution of quantum energy levels—quantified by the Brody parameter—we diagnose transitions between insulating (localized) and metallic (chaotic) states. Our findings reveal that tungsten impurity concentration, gate voltage, and temperature serve as interdependent control parameters for quantum chaos, enabling deliberate tuning of the channel's metal–insulator transition. In addition, a trained multilayer perceptron accurately predicts the quantum chaotic signature, providing a fast and reliable surrogate for device simulations. {\color{red} This work establishes a new paradigm where machine learning serves as a high-fidelity surrogate for quantum chaos diagnostics, enabling rapid exploration of multidimensional design spaces and inverse engineering of 2D transistors based on fundamental quantum-state criteria rather than empirical performance metrics.

Abstract Rare-earth Barium Copper Oxide (REBCO) high-temperature superconducting (HTS) magnets are typically fabricated by winding REBCO tapes into coils. During operation, these magnets experience compressive stresses induced by winding tension, electromagnetic forces, and thermal contraction, which can cause mechanical deformation and fracture. Conventional approaches estimate the mechanical properties of REBCO tapes based on the behavior of individual constituent materials; however, such estimates are valid only for single tapes. In practical applications, stacked or wound tapes exhibit markedly different mechanical responses due to interfacial effects. Therefore, investigating the mechanical behavior of stacked HTS tapes is essential for reliable structural design and analysis. In this study, five types of REBCO tapes were selected and cut into uniform dimensions to fabricate stacked test specimens. Compression tests were conducted in a liquid-nitrogen-filled vacuum chamber at two maximum stress levels to simulate typical stresses experienced during magnet cool-down and operation. Two stacking configurations were considered: one consisting solely of REBCO tapes and the other composed of alternating REBCO and stainless-steel tapes, commonly referred to as metal insulation (MI). Precision cutting tools were employed to minimize edge effects, and custom specimen guides and load transfer blocks were designed to ensure proper transversal loading. A cryogenically compatible universal testing machine (UTM) was used to apply compressive stress, while an extensometer measured the displacement. Each specimen underwent three loading-unloading cycles to evaluate hysteresis behavior after local post-yield. Stress-strain curves were obtained, and the effective tangent modulus was determined. Under maximum compressive stress, the measured modulus varied significantly depending on the specimen, ranging from 2.5 GPa to 20.3 GPa, and exhibited clear stress-dependent behavior. The copper stabilizer thickness ratio, associated with low stiffness, strongly influenced both the strain and tangent modulus. These results provide key mechanical properties necessary for the structural evaluation and modeling of HTS magnets.

Wideband tunable terahertz metamaterial absorber based on vanadium dioxide (VO2) for terahertz sensing and communication