Search Results (342)

The quantitative characterization of near-surface heterogeneity using ground-penetrating radar (GPR) is an important but challenging task. The estimation of subsurface geostatistical parameters from surface-based common-offset GPR reflection data has so far relied upon a Monte-Carlo-type inversion approach. This allows for a comprehensive exploration of the parameter space and provides some measure of uncertainty with regard to the inferred results. However, the associated computational costs are inherently high. To alleviate this problem, we present an alternative deep-learning-based technique, that, once trained in a supervised context, allows us to perform the same task in a highly efficient manner. The proposed approach uses a convolutional neural network (CNN), which is trained on a vast database of autocorrelations obtained from synthetic GPR images for a comprehensive range of stochastic subsurface models. An important aspect of the training process is that the synthetic GPR data are generated using a computationally efficient approximate solution of the underlying physical problem. This strategy effectively addresses the notorious challenge of insufficient training data, which frequently impedes the application of deep-learning-based methods in applied geophysics. Tests on a wide range of realistic synthetic GPR data generated using a finite-difference time-domain (FDTD) solution of Maxwell’s equations, as well as a comparison with the results of the traditional Monte Carlo approach on a pertinent field dataset, confirm the viability of the proposed method, even in the presence of significant levels of data noise. Our results also demonstrate that typical mismatches between the dominant frequencies of the analyzed and training data can be readily alleviated through simple spectral shifting. Full article

Laser-induced periodic surface structures (LIPSSs) produced via ultrafast laser-induced oxidation offer a promising route for high-quality nanostructuring, with reduced thermal damage compared to conventional ablation-based methods. However, the influence of laser repetition frequency on the formation and morphology of oxidized LIPSSs remains insufficiently explored. In this study, we systematically investigate the effects of varying the femtosecond laser repetition frequency from 1 kHz to 100 kHz while keeping the total pulse number constant on the oxidation-induced LIPSSs formed on amorphous silicon films. Scanning electron microscopy and Fourier analysis reveal a transition between two morphological regimes with increasing repetition frequency: at low frequencies, the long inter-pulse intervals result in irregular, disordered oxidation patterns; at high frequencies, closely spaced pulses promote the formation of highly ordered, periodic surface structures. Statistical measurements show that the laser-modified area decreases with frequency, while the LIPSS period remains relatively stable and the ridge width exhibits a peak at 10 kHz. Finite-difference time-domain (FDTD) and finite-element simulations suggest that the observed patterns result from a dynamic balance between light-field modulation and oxidation kinetics, rather than thermal accumulation. These findings advance the understanding of oxidation-driven LIPSS formation dynamics and provide guidance for optimizing femtosecond laser parameters for precise surface nanopatterning. Full article

Optically pumped gas terahertz (THz) lasers (OPGTLs) as reliable sources of THz radiation have been extensively utilized within THz application areas. In this paper, a substrate-free metal mesh coupler based on the metasurfaces principle was designed for continuous wave OPGTL, which is suitable for the Fabry–Perot (FP) THz resonator. The parameters of substrate-free metal mesh are calculated by the Ulrich equivalent circuit model, and the influence of metal mesh period and linewidth on its transmittance is analyzed quantitatively. Taking the THz laser with the 118.8 µm of CH₃OH optically pumped by the 9.6 µm CO₂ laser line for instance, two kinds of metal mesh were devised as input and output couplers of the resonator, and the transmittance and reflectance of the metal meshes are verified by the finite-difference time-domain (FDTD) method. Furthermore, the transmitted and reflected light fields of the FP resonant cavity metal mesh mirrors were simulated by using the FDTD method under the vertical incidence of both pump light and THz waves. Validation of the optical field characteristics of the substrate-free metal meshes confirmed their suitability as ideal input and output coupling cavity mirrors for FP resonant cavities in optically pumped gas THz lasers. Full article

Nickel (Ni) ultrathin films exhibit phase-dependent electrical, magnetic, and optical characteristics that are significantly influenced by deposition methods. However, these films are inherently prone to rapid oxidation, with the oxidation rate dependent on substrate, temperature, and deposition parameters. The focus of this research is to investigate the temporal oxidation of RF-sputtered Ni ultrathin films on Corning glass under ambient atmospheric conditions and its impact on their structural, surface, and optical characteristics. Controlled film thicknesses were achieved through precise manipulation of deposition parameters, enabling the analysis of oxidation-induced modifications. Atomic force microscopy (AFM) revealed that films with high structural integrity and surface uniformity are exhibiting roughness values (Rq) from 0.679 to 4.379 nm of corresponding thicknesses ranging from 4 to 85 nm. Scanning electron microscopy (SEM) validated the formation of Ni grains interspersed with NiO phases, facilitating SPR-like effects. UV-visible spectroscopy is demonstrating thickness-dependent spectral (plasmonic peak) shifts. Finite Difference Time Domain (FDTD) simulations corroborate the observed thickness-dependent optical absorbance and the resultant shifts in the absorbance-induced plasmonic peak position and bandgap. Increased NiO presence primarily drives the enhancement of electromagnetic (EM) field localization and the direct impact on power absorption efficiency, which are modulated by the tunability of the plasmonic peak position. Our work demonstrates that controlled fabrication conditions and optimal film thickness selection allow for accurate manipulation of the Ni oxidation process, significantly altering their optical properties. This enables the tailoring of these Ni films for applications in transparent conductive electrodes (TCEs), magneto-optic (MO) devices, spintronics, wear-resistant coatings, microelectronics, and photonics. Full article

(1) Background: Asomate, as a dithiocarbamate compound, is moderately toxic to the human body; thus, it is necessary to develop a rapid and efficient method for detection. To meet this need, this study introduced a rapid, non-destructive, and efficient method for detecting asomate residues on the surface of apples based on surface-enhanced Raman spectroscopy (SERS) combined with flexible substrates. (2) Methods: Concave Au nanorods (AuCNAs) were synthesized in advance. Then, the AuCNAs were loaded on an electrostatically spun film to generate a flexible TiO₂/ZrO₂/AuCNAs substrate for detection. (3) Results: The flexible substrate exhibited strong SERS activity, with an enhancement factor (EF) up to 9.40 × 10⁷ for 4-MBA. Meanwhile, the finite-difference time-domain (FDTD) simulation showed that the localized surface plasmon resonance (LSPR) effects related to the enhancement of the SERS signal are mainly generated from the ‘hot spots’ in AuCNAs. The density functional theory (DFT) simulation detailedly revealed that the SERS peaks could be generated by the interaction among asomate molecules, disassociated Au atoms, and Au facets. Moreover, the asomate in apple peel was analyzed with the limit of detection (LOD) as low as below 10 nM, allowing for the rapid detection of asomate directly on apple peels. (4) Conclusions: The flexible TiO₂/ZrO₂/AuCNAs film can be used for the in situ detection of asomate in apple peel at low concentrations. Moreover, the simulation methods, including FDTD and DFT, explained the mechanism of SERS from the flexible substrates. Full article

The continuous demand for faster processing systems, driven by the rise of artificial intelligence, has exposed limitations in traditional transistor-based electronics, including quantum tunneling, heat dissipation, and switching delays due to challenges in further miniaturization. This study explores optical systems as a promising alternative, leveraging the speed of photons over electrons. Specifically, we design and simulate optical NAND and NOR logic gates using a two-dimensional photonic crystal structure with a square lattice. Symmetrical waveguides are used for the input paths to make the structure relatively more straightforward to fabricate. A key innovation is the ability to realize both gates within a single structure by adjusting the phases of the input sources. To optimize the phase parameters efficiently, we employ the ML-FOLD (Meta-Learning and Formula Optimization for Logic Design) optimization formula, which outperforms traditional methods and machine learning approaches in terms of computational efficiency and data requirements. Through finite-difference time-domain (FDTD) simulations, the proposed optical structure demonstrates successful implementation of NAND and NOR gate logic, achieving high contrast ratios of 4.2 dB and 4.8 dB, respectively. The results validate the effectiveness of the ML-FOLD method in identifying optimal configurations, offering a streamlined approach for the design of all-optical logic devices. Full article

Solar photovoltaic/thermal (PV/T) systems can simultaneously solve PV overheating and obtain high-quality thermal energy through nanofluid spectral splitting technology. However, the existing nanofluid splitting devices have insufficient short-wavelength extinction and stability defects. To achieve the precise matching of the nanofluid splitting performance with the optimal spectral window of the PV/T system, this paper carries out a relevant study on the optical properties of Al nanoparticles and proposes an Al@Ag nanoparticle. The optical behaviors of nanoparticles and nanofluids are numerically analyzed using the finite-difference time-domain (FDTD) method and the Beer–Lambert law. The results demonstrate that adjusting particle size enables modulation of nanoparticle extinction performance, including extinction intensity and resonance peak range. The Al@Ag core–shell structure effectively mitigates the oxidation susceptibility of pure Al nanoparticles. Furthermore, coating Al nanoparticles with an Ag shell significantly enhances their extinction efficiency in the short-wavelength range (350–640 nm). After dispersing Al nanoparticles into water to form a nanofluid, the transmittance in the short-wavelength range is significantly reduced compared to pure water. Compared to 50 nm pure Al particles, the Al@Ag nanofluid further reduces the transmittance by up to 13% in the wavelength range of 350–650 nm, while having almost no impact on the transmittance in the photovoltaic window (640–1080 nm). Full article

Lightning-induced voltages on overhead distribution lines present a formidable obstacle to ensuring the reliability of power systems, evaluated through conventional numerical techniques, such as the Finite Difference Time Domain (FDTD) method and the Finite Element Time Domain (FETD) method. This study proposes a novel implementation of the Radial Basis Function-Finite Difference Time Domain (RBF-FDTD) method, extending the foundation of our previous work to address the field-to-line coupling equations governing such systems. The effectiveness and accuracy of this approach are rigorously validated through RBF-FDTD numerical simulations, applied to both horizontal and vertical configurations of a 1 km, 110 kV multi-conductor distribution line, as well as a real-world three-phase overhead line in Vietnam. In this study, the impact of various parameters, including line geometry, the presence of ground wires, and the influence of perfectly and imperfectly conducting ground, on the lightning-induced voltages are investigated. The simulation and computational results are in good agreement with findings from prior studies, underscoring the potential of the RBF-FDTD method as a robust tool of practical implications. Full article

The electric field enhancement effect induced by localized surface plasmon resonance (LSPR) plays a critical role in imaging and sensing applications. In particular, nanocube structures with narrow gaps provide large hotspot areas, making them highly promising for high-sensitivity applications. This study predicts the electric field enhancement effect of structures combining silver nanocubes and a 10 nm thick silver thin film using the finite-difference time-domain (FDTD) method. We demonstrate that the interaction between the silver nanocubes and silver thin film allows control over sharp LSPR peaks in the visible wavelength range. Specifically, the structure with a spacer layer between the silver nanocubes and the silver thin film is suitable for multimodal imaging, while the direct contact structure of the silver nanocubes and the silver thin film shows potential as a highly sensitive refractive index sensor. The 10 nm thick silver thin film enables backside illumination due to its transparency in the visible wavelength region, making it compatible with inverted microscopes and allowing for versatile applications, such as living cell imaging and observations in liquid media. These structures are particularly expected to contribute to advancements in bioimaging and biosensing. Full article

Extracting defect profile parameters from measured defect images poses a significant challenge in extreme ultraviolet (EUV) multilayer defect metrologies, because these parameters are crucial for assessing defect printing behavior and determining appropriate repair strategies. This paper proposes to reconstruct defect profile parameters from reflected field intensity images of a phase defect assisted by transfer learning with fine-tuning. These images are generated through simulations using the rigorous finite-difference time-domain (FDTD) method. The VGG-16 pre-trained model, known for its robust feature extraction capability, is adopted and fine-tuned to map the intensity images to the defect profile parameters. The results demonstrate that the proposed approach accurately reconstructs multilayer defect profile parameters, thus providing important information for mask repair strategies. Full article

In computational electromagnetics, the finite-difference time-domain (FDTD) method is recognized for its volumetric discretization approach. However, it can be computationally demanding when addressing large-scale electromagnetic problems. This paper introduces a novel approach by incorporating Graphic Process Units (GPUs) into an FDTD algorithm. It leverages the Compute Unified Device Architecture (CUDA) along with OpenMPI and the NVIDIA Collective Communications Library (NCCL) to establish a parallel scheme for the FDTD algorithm in distributed cluster GPUs. This approach enhances the computational efficiency of the FDTD algorithm by circumventing data relaying by the CPU and the limitations of the PCIe bus. The improved efficiency renders the FDTD algorithm a more practical and efficient solution for real-world electromagnetic problems. Full article

Clinical research groups rarely have easy access to dedicated animal Magnetic Resonance (MR) systems. For this reason, dedicated hardware has to be developed to optimize small animal imaging on clinical scanners. In MR systems, radiofrequency (RF) coils are key components in the acquisition process of the MR signal, and the design of hand-crafted, organ-specific RF coils can be a constraint in many research projects. Accurate design and simulation processes enable the optimization of RF coil performance for a given application by avoiding trial-and-error approaches. This paper describes the full-wave simulation of a solenoidal coil for Magnetic Resonance Imaging (MRI) using the finite-difference time-domain (FDTD) method. Such a simulator enables the estimation of the coil’s magnetic field pattern in a loaded condition, the coil inductance, and the sample-induced resistance. The resulting accuracy is verified with data acquired with a solenoid prototype designed for small animal experiments with a 3T MRI clinical scanner. Full article

The Finite Difference Time-Domain (FDTD) method is a powerful tool for electromagnetic field analysis. In this work, we develop a variation of the algorithm to accurately calculate antenna, microwave circuit, and target scattering problems. To improve efficiency, a single-field (SF) FDTD method is proposed as a numerical solution to the time-domain Helmholtz equations. New formulas incorporating resistors and voltage sources are derived for the SF-FDTD algorithm, including hybrid implicit–explicit and weakly conditionally stable SF-FDTD methods. The correctness of these formulas is verified through numerical simulations of a newly designed dual-band wearable antenna with an artificial magnetic conductor (AMC) structure. A novel antenna fed by a coplanar waveguide with a compact size of 15.6 × 20 mm² has been obtained after being optimized through an artificial intelligent method. A double-layer, dual-frequency AMC structure is designed to improve the isolation between the antenna and the human body. The simulation and experiment results with different bending degrees show that the antenna with the AMC structure can cover two frequency bands, 2.4 GHz–2.48 GHz and 5.725 GHz–5.875 GHz. The gain at 2.45 GHz and 5.8 GHz reaches 5.3 dBi and 8.9 dBi, respectively. The specific absorption rate has been reduced to the international standard range. In particular, this proposed SF-FDTD method can be extended to analyze other electromagnetic problems with fine details in one or two directions. Full article

In this study, we propose a new nonlinear polarization model that modifies the polarization equation to account for the material’s nonlinear response. Specifically, the nonlinear restoring force in our model is reformulated as an electric field-dependent function, derived from the nonlinear Lorentz model. Additionally, we perform a comparative analysis of the Kerr model, the Duffing model, the nonlinear Lorentz model, and our modified nonlinear Lorentz model (MNL) by solving Maxwell’s equations using the finite-difference time-domain (FDTD) method. This research focuses on the third-order nonlinearity of these models under varying light intensities and different ratios of resonant frequency to carrier frequency. First, in the example we studied, our results show that the MNL model produces results closer to the Kerr model when the light intensity is significantly high. Second, the comparison under different resonant frequencies reveals that all models converge to the Kerr model when the carrier frequency is much lower than the resonant frequency. However, when the carrier frequency significantly exceeds the resonant frequency, the differences between the Kerr model and the other models become more noticeable. The third-order nonlinearity of our MNL model aligns more closely with the Kerr model than the nonlinear Lorentz and Duffing models do when the ratio of resonant frequency to carrier frequency is between 1 and 2. Full article

Numerical simulation of ground-penetrating radar (GPR) has been widely used to enhance the interpretation of GPR data and serves as a key component in Full Waveform Inversion (FWI). In response to the time-consuming numerical computation of layered medium and buried targets, which leads to inefficiency in full-wave inversion, this paper proposes a machine learning-based forward scattering rapid solution method. Using the detection of rebar buried in concrete under sand as the GPR application scenario, with scene parameters such as concrete moisture content, rebar radius, and burial depth, scattering echo signals are obtained via Finite Difference Time Domain (FDTD) simulation. Principal component analysis (PCA) is applied to reduce the dimensionality of the echo data, and the first 40 principal component weight coefficients are selected as the output of the deep learning network. An innovative cyclic nested deep learning network architecture is designed, which not only fully explores the intrinsic causal relationship between the scene parameters and the principal component weight coefficients, but also refines and corrects each predicted principal component. The numerical results demonstrate that, compared with traditional machine learning methods, the cyclic nested machine learning network architecture offers higher prediction accuracy and learning efficiency, validating the effectiveness of the proposed method. Full article