Due to their excellent physical properties, InAs/InxGa1-xAsySb1-y type-II superlattices (SLs) have become promising materials for the active regions of infrared devices. Their translation into practical applications, especially in harsh environments such as space, requires detailed knowledge of their structural and optoelectronic stability under extreme conditions. The response of these quaternary SLs to high-energy electron irradiation, however, has not been extensively investigated. In this work, the structural and luminescence properties of InAs/In0.5Ga0.5As0.5Sb0.5 SLs exposed to 1.25 MeV electron irradiation at absorbed doses of 0, 5, 50, and 100 kGy are investigated. High-resolution X-ray diffraction (HRXRD), atomic force microscopy (AFM), and transmission electron microscopy (TEM)/energy-dispersive X-ray spectroscopy (EDS) measurements show irradiation-induced peak broadening, increased lattice mismatch, surface roughening, interface blurring, As/Sb interdiffusion, and progressive In redistribution, while the overall superlattice periodic stacking remains observable. Geometric phase analysis (GPA) further reveals dose-dependent modification of the local strain distribution. Power-dependent photoluminescence (PL) measurements indicate a dose-dependent shift from predominantly free-exciton-like radiative recombination toward an increasing contribution from defect-assisted non-radiative recombination. These results establish a correlation between irradiation dose, interfacial and compositional disorder, carrier localization, and optical degradation in the quaternary type-II superlattice.
The chemical and microstructural stability of proton-conducting electrolyte is crucial for durable operation of protonic ceramic cells (PCCs). In this study, the reduction stability of BaZr0.1Ce0.7Y0.1Yb0.1O3−δ (BZCYYb) was systematically investigated. At 400-800 °C, wet H2 caused more severe surface precipitation on BZCYYb than dry H2. Exposed to humidified H2 at 800 °C, Ba depletion and Ce4+ reduction promoted surface segregation of Zr, Y, Ce, and Yb and the formation of a 15-20 nm oxide layer on the surface of BZCYYb. Compared to the fresh BZCYYb, the conductivity of BZCYYb reduced in dry and wet H2 at 800 °C decreased, while the conductivity of BZCYYb reduced in wet H2 at 400-600 °C tended to increase. The segregation of multiple oxides from BZCYYb complicated the conductivity change trend. Tested at 200 °C in moist air, the total conductivities of BZCYYb reduced in moist H2 for 50h at 400 °C, 600 °C, and 800 °C were respectively 6.5E-5, 6.3E-5, and 1.8E-6 S cm−1, while those were 4.9E-5, 2.9E-5, and 1.9E-5 S cm−1 reduced in dry H2. These findings disclose the origins inducing the instability of the electrolyte in reducing atmosphere, provide insight into degradation mechanisms of proton-conducting electrolytes, and lay the foundation for design of stable electrolyte under practical operation conditions.
Multivariate time series classification (MTSC) serves as a fundamental task across multiple domains such as finance and healthcare. For different MTSC tasks, the discriminative features may either manifest as multi-scale temporal features or reside in different frequency components. While current approaches predominantly rely on single-view pattern, significantly limiting their generalizability. To address this issue, we propose a novel multi-view learning and mixing framework (TSCMixer) that generates more comprehensive feature representations, enhancing the network’s robustness to scale and spectral distribution of discriminative features. The framework implements multi-view learning by: multi-scale temporal convolutions for hierarchical pattern extraction, and discrete wavelet transform (DWT) for time-frequency analysis, separating high-frequency (detail coefficients) and low-frequency (approximation coefficients) components. These components are processed through parallel Mamba-based streams, effectively capturing long-range dependencies. To mitigate isolation between high- and low-frequency features, we design adaptive frequency bridges that enable spectral mixing, enhancing feature fusion. Additionally, dual attention mechanisms further enhance the feature representations. Finally, the multi-scale temporal features, high-frequency and low-frequency features in time-frequency domain are fused through cross-domain concatenation and processed by a lightweight classifier to predict the classification labels. Extensive experiments on 16 UEA benchmark datasets show our method outperforms 16 state-of-the-art methods, achieving an average accuracy of 83.6%. Code is available at: https://github.com/HuiyingLiu111/TSCMixer.
Engineering semiconductor heterojunctions is a pivotal research frontier for developing next-generation optoelectronic devices, particularly for neuromorphic computing. However, performing precise, stimulus-selective control over interfacial carrier dynamics to unlock advanced memory and logic functions remains a significant challenge. Herein, we unveil a novel "photon-energy-gated selective carrier accumulation" mechanism within macroscopically ordered ZnO/Ga2O3 core-shell nanowire heterojunctions, governed by the Type-I band alignment at the heterointerface. This unique architecture, fabricated via a super-aligned carbon nanotube template, not only boosts the UV photoresponsivity by 78-fold compared to pure ZnO counterparts but also enables a remarkable wavelength-selective activation of persistent photoconductivity. Harnessing this optically switchable memory effect, we demonstrate its application in optoelectronic synaptic devices, and we also prove its potential in physical reservoir computing system, achieving a high classification accuracy of 87.9% on the Fashion - Modified National Institute of Standards and Technology (F-MNIST) dataset. This study provides a new paradigm for designing intelligent optoelectronic devices by precisely manipulating interfacial carrier dynamics, opening promising avenues for in-sensor computing and advanced photonic memory.
Underwater laser cleaning has emerged a promising approach for marine infrastructure maintenance; however, the underlying material removal mechanisms remain insufficiently understood. In this study, nanosecond pulsed laser cleaning (1064 nm, 100 ns, and 20 kHz) of a biofouling layer on stainless steel submerged in seawater was investigated through a combined finite element model and experimental approach. The simulation results showed that the transient temperature rise remained localized near the surface, reaching a peak temperature of 594.47 K at a fluence of 0.07 J/cm2 within the biofouling layer. The computed near-surface thermoelastic stress of 1.33 MPa substantially exceeded the reported adhesion strength of soft marine biofouling (5–200 kPa). Experimental validation using FTIR, SEM, and AFM was used for surface characterization. The findings indicate an effective removal of biofouling within a low fluence of 0.04–0.06 J/cm2, achieving near-complete cleaning with improved surface smoothness (Ra ∼ 127 ± 1.4 nm). On the other hand, at higher fluence, the surface roughness increased, and fine micro-cracks developed, which indicates the beginning of thermomechanical surface modification. These results indicate that thermoelastic stress is the dominant removal mechanism under low fluence conditions and provide quantitative parameters for underwater laser cleaning processes. The observed trending in morphology introduces a quality-based parameter that can be used in the selection of underwater laser cleaning parameters to achieve efficient fouling removal without substrate damage.
