Journal of Computational Electronics最新文献

We propose a series of derivatives (HDXR–HDX7) using the xanthenone-based photoactive compound (HDX) by introducing various electron-withdrawing groups at its terminal acceptors, emphasizing their potential as photovoltaic materials. Density functional theory (DFT) and time-dependent DFT (TD-DFT) calculations were employed using the M06 functional along with the 6-311G(d,p) basis set to investigate the optoelectronic and photovoltaic properties of these compounds. The frontier molecular orbitals (FMOs) revealed their energy gap in the range of 2.859−3.378 eV with the following trend: HDX4 < HDX3 < HDX6 < HDX5 < HDX7 <HDX2 < HDX1 < HDXR. The global reactivity parameters (GRPs) calculated through highest occupied molecular orbital (HOMO)/lowest unoccupied molecular orbital (LUMO) energies illustrated higher softness with reduced hardness values. All designed derivatives exhibited redshifted absorption spectra (446.903–516.407 nm) in comparison with HDXR (444.276 nm) and revealed enhanced light-harvesting efficiency. Moreover, the density of states (DOS) plots showed significant transfer of charges between HOMO and LUMO. The lower binding energy (E_b) values (0.458–0.594 eV) exhibited higher exciton dissociation strength, which also supported their transition density matrix (TDM) maps. Among all the designed derivatives, HDX4 showed the most suitable results for solar cells, with a minimal energy gap of 2.859 eV and E_b = 0.458 eV along with the highest bathochromic shift (λ_max = 516.407 nm). The photovoltaic properties, in particular the open-circuit voltage (V_oc), were calculated with respect to the PBDB-T donor, and comparable results were obtained for all the investigated chromophores. Thus, the results indicate that the proposed xanthenone-based chromophores are good candidate materials for use in photovoltaic applications.

This work presents the theoretical investigation of a one-dimensional defective photonic crystal structure proposed for edible oil impurity detection.The structure consists of alternating MgF₂ and TiO₂ layers with different oil mixtures introduced as the defect layer. Mustard oil is considered as the base oil, while argemone oil, rice bran oil, and palm oil are treated as impurity oils. The transmission characteristics of the proposed structure are investigated using the Transfer Matrix Method (TMM). The results show that the defect mode wavelength shifts systematically with changes in impurity concentration and refractive index. The effect of periodic layers and defect layer thickness on sensor performance is also analyzed. The proposed sensor exhibits high transmittance, narrow resonance linewidth, high quality factor, and excellent figure of merit. A sensitivity of nearly 298–299 nm/RIU, Q-factor- 3.6 × 10⁵, figure of merit (FOM) 6.0 × 10⁴ (RIU⁻¹) and a detection limit of 1.64 × 10^–6 RIU are achieved, indicating the excellent sensing capability of the proposed structure for edible oil quality monitoring applications.

This study investigates the improvement in retention characteristic in the reprogram scheme of 3D NAND Flash memory using a physical model. The silicon-oxide-nitride-oxide-silicon (SONOS) structure is employed, and the charge-trap layer (CTL) is analyzed according to trap levels. Traps inside the CTL are categorized into shallow trap levels (STL) and deep trap levels (DTL), with multiple levels sampled according to the distribution of trap concentration as a function of trap depth. Charge loss during the reprogram interval is analytically calculated based on electron emission, and the relative contributions of shallow and deep traps are compared after the second program operation. The results demonstrate that the electron trap distribution shifts toward a higher DTL concentration compared with the one-shot program. Furthermore, a trade-off relationship is proposed, in which retention characteristic can be tuned by adjusting the interval time between two program operations. The results were calibrated against measurement data and validated with technology computer-aided design (TCAD) simulations, and a voltage biasing scheme is proposed to enhance reliability.

Recently, plasmonic photonic crystal fiber (PCF)-based refractive index sensors have drawn interest for their high sensitivity and capability to perform label-free, real-time detection. Therefore, this paper describes the design and functionality of a new D-shaped PCF biosensor combined with black phosphorus (BP)–gold and Ta₂O₅ layers for refractive index sensing and early detection of anemia. To achieve light confinement in PCF, this design includes two kinds of air holes: elliptical and circular. The proposed sensor is coated with a layer of gold and a Ta₂O₅ adhesion layer, which serves a dual purpose by enhancing the plasmonic response while simultaneously providing adhesion between the PCF and the gold coating. Additionally, 17 layers of BP—a growing 2D plasmonic material—are also layered on the gold + Ta₂O₅ layer to enhance sensitivity, which has a monolayer thickness of 0.53 nm. The proposed sensor is used to detect a wide range of RI ranging from 1.31 to 1.38. After incorporating BP layers, the dynamic sensitivity of developed sensor enhanced from 5500 to 12,000 nm/RIU. The BP-layered hybrid sensor achieved amplitude sensitivity of − 252.28 RIU⁻¹. The proposed structure is numerically analyzed with the finite element method (FEM) in COMSOL Multiphysics. Furthermore, the optimized structure achieved wavelength sensitivity of 5500 nm/RIU at 140 g/L for the gold-coated PCF sensor and 14,500 nm/RIU at 140 g/L for BP–gold-coated PCF sensor for anemia detection. Therefore, the suggested PCF-SPR biosensor based on black phosphorus could be useful for the early diagnosis of anemia.

This work presents a phase-accumulation-based machine learning readout for dynamically perturbed graphene terahertz resonators. Instead of relying on steady-state or peak-only observables, the proposed approach integrates the transient phase response to construct physically meaningful features that capture both perturbation magnitude and temporal behavior. A physics-driven synthetic dataset is generated using a reduced-order dynamic model, incorporating parameter variability and noise. Multiple feature configurations are evaluated for classification and regression tasks. The results show that accumulated-phase descriptors alone are insufficient, but significantly improve performance when combined with instantaneous features, leading to enhanced robustness and predictive accuracy, particularly under low signal-to-noise conditions. The proposed framework provides a physics-informed pathway for integrating machine learning into graphene-based terahertz sensing systems.

Real-time time-dependent density functional theory (TDDFT) constitutes a cornerstone in the modeling of nanoscale materials. Although numerous application areas have already benefited from the technique, investigation of both forward and backward coupling in matter systems subject to a vectorial electromagnetic (EM) perturbation remains challenging. To address this problem, this work proposes a modeling framework leveraging the finite-difference time-domain algorithm for determining the scalar and vector potential from the EM fields, both in the traditional Coulomb gauge as well as the nearly unexplored Lorenz gauge. Using the latter allows for a straightforward calculation of the electromagnetic vector potential, a natural implementation of perfectly matched layers, and a significant reduction in computation time. To describe the pertaining quantum-mechanical (QM) dynamics, a real-space and full-potential formulation of TDDFT has been utilized. Furthermore, a multistep algorithm is leveraged to bridge the gap between the stability limits of the EM and QM subsystems. The method is applied to practical nanowire topologies to demonstrate its versatility. The results exhibit an excellent correspondence between both gauges, while highlighting the superior numerical performance of the Lorenz gauge technique.

We develop a fully quantum-dynamical framework for modeling excitonic effects in group-IV semiconductor heterostructures based on a single-band (1s) tight-binding Hamiltonian. The model incorporates position-dependent band-edge potentials for electron and hole confinement, long-range Coulomb interaction, and external electric fields. Starting from the total Hamiltonian (hat{H}^{tb} = hat{H}^{tb}_0 + hat{H}^{tb}_{em} + hat{H}^{tb}_C), we derive the equations of motion for the density matrix in superoperator form and solve them numerically using an adaptive Runge–Kutta scheme to ensure temporal accuracy for fast field-driven dynamics. This approach enables the calculation of stationary exciton states as well as time-resolved electronic and optical observables in realistic one-dimensional heterostructures, including single and multiple quantum wells. We systematically analyze how excitonic binding energies, absorption spectra, and carrier dynamics depend on Sn composition, quantum well geometry, and external driving fields. While multi-quantum well configurations can be engineered to modify the optical response, our results show that absorption enhancement critically depends on the spatial overlap and Coulomb coupling between confined electron and hole states. The presented framework provides microscopic insight into excitonic processes relevant for the design of Sn-based group-IV photonic materials operating in the mid-infrared regime.

This work presents a physics-based compact model for double-gate junctionless field-effect transistors, with emphasis on accurately capturing the impact of ungated source/drain extensions on the drain current characteristics. The model is validated against two-dimensional device simulations performed using Silvaco ATLAS for two channel doping concentrations and a wide range of ungated extension lengths. To isolate the contribution of the access regions and clarify the effective channel length, all mobility degradation models were disabled in the simulations, allowing the observed current degradation to be attributed solely to the series resistance of the ungated extensions. The proposed formulation includes an analytical factor ξ that accounts for the reduced electrostatic influence of the source and drain terminals on the channel potential, as well as a closed-form expression for the fringe capacitance associated with the ungated regions. The resulting drain current model demonstrates very good agreement with numerical simulations across different geometries and doping levels. Model symmetry is further verified through a Gummel symmetry test, confirming the physical consistency of the formulation. Owing to its analytical nature and physical transparency, the proposed model is well suited to serve as a core building block for higher-level compact models of JL devices.

Self-heating effects (SHEs) pose significant challenges for advanced semiconductor devices, particularly gate-all-around field-effect transistors (GAAFETs). Although finite element simulation and thermal network modeling serve as robust electrothermal analysis tools, their reliance on intricate physical mechanisms and numerous parameters results in prohibitively high computational costs. By contrast, machine learning (ML) techniques provide an innovative paradigm for predicting device behaviors. In this work, a methodology for developing artificial neural network (ANN) models to rapidly integrate nanosheet geometry-dependent electrothermal properties of advanced GAAFETs into device and circuit simulations is proposed. To accurately replicate SHEs-induced I–V and T–V characteristics, the ANN model is optimized through output nonlinearity control and input range management via conversion functions and a logarithmic transformation-based preprocessing scheme. The trained model achieves test accuracy of ~ 98.5% for drain current and ~ 99% for temperature prediction across diverse geometries and supply voltages. These precise predictions demonstrate that ANN-based model effectively captures underlying thermal physics, exhibiting strong capability and generality for circuit-level performance evaluation.

This paper presents a unified framework for estimating the parameters of equivalent circuits of solar photovoltaic (PV) cells and induction machines. The framework is based on a novel metaheuristic algorithm that relies on the hybridization of the Simulated Annealing (SA) and Equilibrium Optimizer (EO) algorithms, referred to as SA-EO, proposed in this work, which integrates the advantages of local exploitation and global exploration to enhance convergence and accuracy. According to the available literature, solar PV cells can be represented using single-, double- and, most accurately, triple-diode equivalent circuits. In a similar way, an induction machine (IM) can be modeled using single- and double-cage equivalent circuits. The estimation of parameters for both solar PV cells and IMs can be performed using either the manufacturer’s data or experimentally recorded measurements. In both cases, the parameters are typically extracted using analytical, numerical, or metaheuristic optimization techniques. Based on the analogy between multi-diode models of solar PV cells and multi-cage models of IMs, this paper introduces a unified parameter estimation framework applicable to all solar cell and IM configurations. The proposed hybrid metaheuristic algorithm is tested on several benchmark models reported in the literature and further validated experimentally. The obtained results confirm that the proposed unified framework offers improved accuracy, faster convergence, and greater robustness compared to existing parameter estimation methods for both solar PV cells and IMs.