Journal of Computational Electronics最新文献

Solar thermal energy collection technologies remain relatively underutilized despite the increasing global emphasis on renewable and environmentally friendly power sources. In this research, a graphene-based perovskite solar absorber (GbPSA) is designed and investigated for enhanced solar-to-thermal energy conversion. The design is also optimized with Artificial Intelligence algorithms. The proposed thermal absorber demonstrates strong absorption across multiple spectral regions, exhibiting absorptance values of 91.55% in the UV band, 88.37% within the visible region, and 76.59% in the infrared spectrum. The GbPSA achieves an average thermal absorptance of 78.84%, and a maximum absorptance of 97.71% at 340 nm, indicating excellent light-capturing efficiency. Under AM1.5 solar irradiation, the structure attains a thermal absorption efficiency of 90.57%, showcasing its suitability for real-world solar exposure. The machine learning optimization is showing 99% prediction accuracy with minimal error rate. The device features a compact configuration, where each unit cell measures 2500 nm in width and depth and 2550 nm in thickness, making it a thin and space-efficient absorber. The wide spectral response suggests that the absorber can effectively harness different forms of solar energy that reach Earth’s surface. Owing to its impressive energy conversion performance, the GbPSA offers significant potential for use in various solar thermal applications, particularly industrial heat generation and other energy-intensive processes.

The electronic properties of armchair graphene nanoribbons (AGNRs) with antidot structures were investigated using quantum transport simulations. Three width families (3P, 3P + 1, and 3P + 2) were considered to examine the impact of introducing a single antidot with different geometries. Antidots of star, cross, and arrow shapes were placed at various positions and spacings within the nanoribbons. The simulations were carried out using the Kwant python package based on a π-orbital tight-binding model combined with the non-equilibrium Green’s function (NEGF) method. The results show that antidot geometry, position, and spacing have a significant effect on the density of states and conductance. Central placement of an antidot introduces a clear transport gap, while edge positions maintain higher transmission. Closely spaced antidots lead to strong localisation and reduced conductance, whereas larger spacing helps restore conductance steps. These findings confirm that structural modification using antidots provides an effective approach to tune the electronic properties of AGNRs for potential applications in graphene-based nanoelectronics devices. These findings confirm that structural modification using antidots provides an effective approach to tune the electronic properties of AGNRs for potential applications in graphene-based nanoelectronics devices. In this study, a combined analysis of antidot shape, spacing, and position is carried out within the same AGNR framework. The simulations show clear quantitative trends, such as stronger suppression of conductance for star and cross shapes compared to arrow shapes, and a noticeable recovery of conductance when the spacing between antidots is increased.

This study investigates the finite-temperature-dependent electronic transport properties of two double-layer systems (DLS), namely a monolayer–monolayer graphene (MLG-MLG) and monolayer graphene–two-dimensional electron gas (MLG-2DEG), using the Boltzmann transport equation. The conductivity of the systems is examined with respect to the influence of various parameters, including the relative carrier concentration, defined as the ratio of carrier concentration ({(n}^{left(text{c}right)})) to Coulomb impurity concentration ({(n}^{left(text{CI}right)})), short-range impurity concentration arising from point defects and long-range (Coulomb charge) impurity concentration, the relative dielectric constant (({varepsilon }_{text{r}})), defined as the ratio of the dielectric constant of the spacer material (({varepsilon }_{2})) to that of the substrate (({varepsilon }_{3})), and the interlayer distance ((d)). The results indicate a single phase transition point in the MLG-MLG system, while the MLG-2DEG system with the addition of impurities reveals two phase transition points. The conductivity as a function of interlayer distance exhibits opposite behavior in the case of the MLG-MLG and MLG-2DEG. The absence of short-range impurities improves the conductivity, and suitable selection of relative dielectric constants, relative carrier concentration, and interlayer distance can enhance the conductivity of the DLS.

Conventional hole-selective layers (HSLs), such as Spiro-OMeTAD, remain popular in perovskite solar cells (PSCs) but are associated with several disadvantages, including parasitic light absorption, low moisture resistance, and high fabrication cost. To address these issues, we proposed five novel self-assembled monolayer (SAM)-derived donor-p-acceptor HSLs (SDA1-SDA5) to enhance charge transport and energy localization in inverted PSCs. SDA2 and SDA3 were the best performers among them. SDA2 had the lowest hole reorganization energy (0.00768 eV) and the highest hole transfer rate (6.27×10¹⁶ s⁻¹) compared to the reference R7B, making it easier to access the charge. It is demonstrated that they display good HOMO alignment with the perovskite valence band, good light-harvesting capacity with red-shifted absorption to 720 nm, and better solubility and processability suggested by high dipole moments (15.4946D) and negative solvation energies (− 8.89 kcal/mol). The HOMO alignments favor the extraction of holes. The charge separations and the reduced recombination losses were evidenced by a high D-index (6.26 Å), low overlap integrals, and the obvious separation of electrons and holes as seen through the charge density and transition density matrix analyses. This work presents a forward-looking computational approach for developing high-performance HSLs in future PSC technologies.

Graphical Abstract

A one-dimensional photonic crystal optical biosensor is proposed for non-invasive anemia diagnosis by introducing a blood layer as a defect in the crystal structure. The operating principle relies on the dependence of blood refractive index on hemoglobin concentration. Variations in hemoglobin levels alter the refractive index of the defect layer, causing a shift in the defect-mode transmission peak within the photonic band gap, enabling accurate anemia detection. Sensor performance is numerically analyzed using the transfer matrix method, considering sensitivity, sensing efficiency, quality factor, and detection limit. The proposed biosensor achieves a sensitivity of 496 nm/RIU, a quality factor of 1449, and a detection limit of 1.32 × 10⁻³ RIU.

This study presents a ring core photonic crystal fiber (RC-PCF) capable of supporting 290 orbital angular momentum (OAM) modes in the C and L telecommunication bands. The fiber incorporates a spiral air-hole geometry with a varying pattern that generates a gradient refractive index profile, enabling exceptional mode confinement. A high-index chalcogenide (As₂S₃) ring core surrounded by silica (SiO₂) cladding enables stable propagation with > 95% mode purity and minimal inter-channel crosstalk, ensured by an effective refractive index difference exceeding 10^–4 for all modes. Structural optimization yields a large numerical aperture (0.33–0.36), ultra-low confinement losses (10^–10–10^–12 dB/m), and a low nonlinear coefficient (0.19–0.23 W⁻¹ km⁻¹). Bending analysis confirms strong tolerance, while 2π and 10 ps walk-off lengths validate signal integrity under practical conditions. The proposed chalcogenide-silica PCF combines low nonlinearity, high mode density and improved bend tolerance making it a revolutionary platform for high-capacity space-division multiplexing in next-generation optical networks.

Persistent divalent metal cations such as Cu²⁺ and Mg²⁺ accumulate in freshwater systems and pose measurable risks to water quality and human exposure. Established analytical techniques, including inductively coupled plasma mass spectrometry and atomic absorption spectroscopy, require centralized laboratories, extensive sample preparation, and batch processing, which limits their use for continuous in situ monitoring at low concentrations. This study presents a terahertz frequency MXene-based metasurface sensor that integrates graphene, silver, strontium titanate, and copper layers to increase electromagnetic field confinement and coupling between the sensing surface and dissolved ions. The sensing mechanism relies on conductivity changes in MXene nanosheets induced by ion adsorption, modulation of the graphene surface response through electrostatic gating, and refractive index sensitivity in the terahertz band. Finite element simulations show that the sensor achieves a spectral sensitivity of 151.1 GHz/RIU for Cu²⁺ and 227.8 GHz/RIU for Mg²⁺. The corresponding figures of merit are 1145.5 and 1759.5 RIU⁻¹, with theoretical detection limits of 4.7 × 10⁻⁴ RIU and 2.7 × 10⁻⁴ RIU. Resonance frequency shifts exhibit linear relationships with refractive index changes, with R² values above 0.993, and with ion concentration, with R² values above 0.895. Random Forest regression was employed as a surrogate modeling tool to interpolate sensor responses obtained from physics-based simulations. The dataset was divided using an 80:20 train–test split, and model performance was evaluated on unseen test data. High coefficients of determination (R² > 0.998) reflect the smooth, deterministic nature of the simulated electromagnetic response rather than experimental variability.

One of the major reliability challenges in next-generation semiconductor devices includes the self-heating effect (SHE) and gate-induced drain leakage (GIDL). These effects are considered a significant roadblock in scaling; hence, it is necessary to address them. The conventional use of silicon dioxide (SiO₂) in silicon-on-insulator (SOI) has proven effective in reducing leakage currents. However, SiO₂ also tends to impede heat dissipation through the substrate, leading to heat accumulation within the device. Ideally, it is desired that the SOI blocks off the leakage currents selectively, while allowing the flow of heat through the SOI to make the substrate act as a heat sink. The proposed use of boron nitride (BN) as SOI can be a suitable replacement for SiO₂, since it has high thermal conductivity while having high electrical resistivity. Our TCAD simulations show 12 and 6% reductions in lattice temperature for N-type and P-type Forksheet Field-Effect Transistors (FSFETs), respectively. The simulation also shows a significant increase of around 27.42% in the electron mobility when BN is used as SOI instead of SiO₂. The GIDL current has been reduced by two orders of magnitude. Thus, BN as SOI demonstrates better thermal management and the reliability of FSFET. In addition, we have identified and quantified various heating mechanisms: Joule, Peltier, Thomson, and recombination heats, which are responsible for heat generation and thermal hotspots in the device.

Early and non-invasive detection of brain tumors remains a critical challenge in biomedical diagnostics, motivating the development of highly sensitive terahertz (THz) biosensing platforms. In this work, a high-efficiency graphene–copper-based THz metasurface biosensor is proposed for refractive-index-based identification of brain tumors. The sensor architecture uses a nested resonator configuration combined with graphene’s tunable surface conductivity, achieved through chemical potential modulation from 0.1 to 0.9 eV, to enhance electromagnetic field confinement and resonance sensitivity. Numerical analysis demonstrates that the sensor operates effectively over a refractive index range of 1.3333–1.4833 RIU, exhibiting a clear resonance redshift from 0.714 to 0.684 THz. A maximum sensitivity of 1538.462 GHz/RIU and a figure of merit of 15.86 RIU⁻¹ are achieved near an optimal refractive index of 1.3425 RIU. Machine-learning-based regression models are further employed to support predictive performance evaluation, yielding a coefficient of determination (R²) of 0.88. Comparative analysis confirms that the proposed sensor outperforms existing biosensors in terms of sensitivity. This machine-learning-assisted approach enables rapid, robust inference under experimental variability, surpassing conventional simulation models or interpolation methods and highlighting its potential for practical clinical deployment.