Journal of the Korean Physical Society最新文献

We investigate the performance of various deep learning architectures for classifying Beyond the Standard Model (BSM) events against Standard Model (SM) backgrounds using simulated high-energy physics data. Four models were evaluated: a Deep Neural Network (DNN), a Convolutional Neural Network (CNN) on 2D image-like data constructed from tabular data, TabNet, and a hybrid Autoencoder + TabNet pipeline. The data were pre-processed through dimensionality reduction and optimized via Optuna for architecture-specific hyperparameters. All four models achieved high classification performance, with AUC values exceeding 0.998. The CNN and hybrid models performed best in both standard metrics and in signal efficiency vs. background rejection analysis. In particular, the Autoencoder + TabNet model achieved the highest accuracy and F1 score, highlighting the effectiveness of combining unsupervised feature extraction with attention-based decision-making. These results demonstrate the applicability of deep learning, especially hybrid and interpretable architectures, for enhancing event classification in collider-based BSM searches.

This study investigates the effect of annealing temperature from the as-deposited state to 450 °C on the structural, optical, and electrical properties of copper oxide thin films prepared by a simple chemical method. Film thickness varied between 870 and 710 nm, while X-ray diffraction confirmed a monoclinic structure with crystallite sizes in the range of 24–42 nm. The optical band gap decreased from 1.79 eV to 1.29 eV with annealing, accompanied by changes in optical constants such as a refractive index of 2.81 and an absorption index of 1.04 for the film annealed at 350 °C. SEM images showed a clear morphological evolution, with agglomerated clusters at low temperatures gradually transforming into well-defined grains at higher annealing temperatures. XPS confirmed the oxidation state progression from mixed Cu⁺/Cu²⁺ in the as-deposited and 150 °C films to complete Cu²⁺ stabilization above 250 °C, along with reduced surface hydroxyl species at higher annealing. Hall effect measurements showed that the film annealed at 350 °C exhibited the best transport properties, with a minimum resistivity of 2.54 × 10² Ω·cm, maximum conductivity of 3.94 × 10^–3 S·cm⁻¹, hole concentration of 2.39 × 10¹⁸ cm⁻³, Hall coefficient of 2.61 cm³·C⁻¹, and mobility of 1.03 × 10^–2 cm²·V⁻¹·s⁻¹. The electrical and optical improvements at this temperature are ascribed to enhanced crystallinity, oxygen stoichiometry, and film adhesion. In addition, I–V measurements of Ag/CuO/ITO films annealed at 350 °C yielded an ideality factor of 2.39, barrier height of 0.74 eV, photosensitivity of 676.55%, quantum efficiency of 14.24%, responsivity of 11.89 mA/W, and detectivity of 7.52 × 10¹⁰ Jones. These results suggest that 350 °C is the optimal annealing temperature for achieving high-quality CuO thin films suitable for photodetector and solar cell applications.

The aim of this article to investigate the impact of gamma irradiation on photocatalytic efficiency of Co₃O₄ nanoparticles toward degradation of contaminants present in wastewater. For this purpose, as-synthesized Co₃O₄ nanoparticles were exposed to gamma irradiation at three different doses. However, structural and optical parameters were changed by gamma irradiation. Variation in parameters, such as crystalline size, dislocation density, optical band gap, and Urbach energy, can be attributed to crystal defect induced by gamma irradiation. Pl analysis revealed that gamma irradiation produced oxygen vacancies which reduce electron/hole separation and enhanced the light absorption. The main objective of this study, which focused on photocatalytic performance, revealed that Co₃O₄ nanoparticles at the dose of 60 kGy attaining a remarkable efficiency of 96% and 89% of I-Buprofen and Diclofenac Sodium within 140 min. The major factors that influence the degradation performance, including pH effect, scavenging test, and reusability test, were analyzed. Scavenging test manifested that hydroxyl radicals are most reactive species in photodegradation mechanism. In summary, this research suggests that the use of nanomaterials at different gamma doses is an alternate approach to improve the photocatalytic performance.

PLS-II facility accelerates electrons to 3 GeV using a linear accelerator and then injects them into a storage ring for synchrotron light production. The beam quality in the linear accelerator is significantly influenced by the electron source, bunching system. To enhance the existing bunching system in PLS-II, a preliminary investigation into the application of sub-harmonic bunchers was carried out. To analyze and optimize beam performance, beam dynamics simulations were conducted using CST and PARMELA. Two SHB cavity frequencies, 1 GHz and 500 MHz, were considered. An MOGA was utilized to optimize the beam parameters for each SHB configuration. A comparative analysis of beam distributions demonstrated that the use of SHBs could enhance injection efficiency by forming a single main bunch.

The Nikiforov–Uvarov functional analysis method was employed in combination with the Deng–Fan and Eckart potential models to derive energy equations from the solutions of the Schrödinger equation. This study investigates the relationship between the variance of a quantum system and the information-theoretic measures specifically, Fisher information and Shannon entropy in both position and momentum spaces. Variance was calculated from expectation values in conjugate spaces, allowing for the determination of uncertainty products, Fisher information products, and Shannon entropic sums. The findings demonstrate that these information measures satisfy their fundamental inequality bounds: an increase in Fisher information corresponds to reduced uncertainty and higher information content, while a decrease in Fisher information results in increased uncertainty. The Deng–Fan potential model was also applied to compute the energy spectra of CO (X¹Σ⁺) and HCl (X¹Σ⁺) diatomic molecules, with results closely aligning with those reported in the literature through alternative analytical methods. The observed trend of increasing energy with higher principal quantum numbers confirms the quantized nature of energy levels. Furthermore, the interaction potential energy for the (a^{3} sumnolimits_{u}^{ + } {}) state of ⁷Li₂ molecule was modeled. The predicted vibrational energy levels show good agreement with Rydberg–Klein–Rees (RKR) data and previous studies. These results are crucial for understanding molecular spectra, energy transitions, and the dynamics of molecular systems in spectroscopy, chemical reactions, and molecular physics.

This study investigates the thermophysical behavior and heat transfer characteristics of TiO₂/ethylene glycol nanofluid between two horizontal coaxial tubes under the influence of thermal radiation and a magnetic field, with a focus on nanoparticle aggregation effects. A mathematical model incorporating aggregation is utilized and solved numerically using MATLAB’s bvp4c function. The analysis compares scenarios with nanoparticles aggregation (WA) and without nanoparticles aggregation (WOA), examining the impact of key dimensionless parameters: Reynolds number (Re), Hartmann number (Ha), nanoparticles volume fraction (ϕ), radiation parameter (Rd), and Eckert number (Ec). Higher Re enhances velocity, while higher values of Ha suppress velocity. Higher values of nanoparticles volume fraction and Eckert number increase the temperature profile. The Nusselt number, reflecting heat transfer, is consistently higher in the flow model without nanoparticles aggregation. Artificial neural network (ANN) and fuzzy particle swarm optimization (FPSO) are employed to predict Nusselt number, and good precision is seen in the predicted values. It is observed that when Hartmann number (Ha) value is equal to 5, and Reynolds number (Re) changes from 0.8 to 4.8, the Nusselt number values increase by 84.0046% and 83.5241% in the case of nanoparticles aggregation and without nanoparticles aggregation model.

We report on the hybrid dual-emission characteristics of GaN-based ultraviolet light-emitting diodes (UV-LEDs) integrated with CH₃NH₃PbBr₃ (MAPbBr₃) perovskite photoluminescent coatings. The resulting device exhibits two spectrally distinct emission bands: a near-UV peak arising from the InGaN/GaN multiple quantum wells and a green emission generated via radiative reabsorption and subsequent photoluminescence (PL) within the perovskite layer. Contrary to the conventional down-conversion assumption, the green emission intensity diminishes with increasing injection current, while simultaneously exhibiting a substantial spectral narrowing and a fixed red-shift. Both UV and green bands show wavelength shifts and linewidth evolution that indicate complex light–matter interactions at the GaN/sapphire/perovskite interfaces, including nonlinear photon reabsorption, saturation of excitonic recombination, and cavity-modulated optical filtering. These findings redefine the role of perovskite coatings as more than passive converters and suggest their active photonic influence on emission dynamics in hybrid LED architectures.

This study investigates the combined effects of electroosmotic flow (EOF), velocity slip, and induced magnetic field on steady and unsteady electromagnetohydrodynamic (EMHD) Jeffrey nanofluid flow between two vertical plates. Water with dispersed copper (Cu) nanoparticles is considered as the working fluid. The governing nonlinear partial differential equations for velocity, temperature, concentration, and induced magnetic field are solved using the finite difference method (FDM) for the transient case and the method of undetermined coefficients for the steady case. Results show that nanofluid velocity increases with Grashof numbers and permeability, while magnetic fields reduce velocity but enhance induced magnetic fields near the lower wall. Slip conditions increase velocity away from the wall, and higher Prandtl numbers enhance heat transfer. These findings provide insights relevant to microfluidic devices, biomedical drug delivery, and electronic cooling applications.

We explore the energy dependence of the nucleon–nucleon potential for constructing a nucleus–nucleus interaction to better describe the volume integral of the potential in the low-energy region. We review the determination of a pseudo-potential expressed as a delta function with a linear energy dependence in the well-known M3Y potential. We then propose a fractional form as a non-linear representation of this energy dependence, considering the low-energy region of the volume integral per nucleon with various target nuclei. The results are compared to the widely used M3Y potential and experimental data. Finally, we present the potentials and the elastic cross section with both linear and non-linear energy-dependent potentials in a p+(^{208})Pb system at 16 MeV, and compare the calculated cross sections to experimental data.

We investigated the characteristics of blue and white inverted organic light-emitting diodes (IOLEDs) varying indium tin oxide (ITO) thicknesses. We measured the optical transmittance, reflectance, sheet resistance, and surface roughness of ITO films with different thicknesses and performed optical simulations to calculate external quantum efficiencies (EQEs) for the devices. The simulation results indicate that the effect of ITO thickness on device efficiency decreases as the number of emission colors increases. We fabricated blue, 2- and 3-wavelength white IOLEDs. The blue IOLEDs with 70 nm ITO showed a 13% higher EQE compared to that of the device with 150 nm ITO. However, the devices with 70 nm ITO exhibited 6% and 1% higher EQEs compared to those of the device with 150 nm ITO in the 2- and 3-wavelength white IOLEDs, respectively, suggesting that the effect of ITO thickness on the EQE of white IOLEDs is weaker than in single-color IOLEDs.