In this study, SCAPS-1D simulations were used to investigate the influence of TiO₂ on the performance of CdTe-based solar cells, particularly in the presence of defects. The proposed device employs AZO as the front contact, TiO₂ as the electron transport layer, CdTe as the absorber, and Ni as the back contact. The simulations were carried out under AM1.5G illumination at 300 K for defective cases. The CdTe thickness (1.5–4 μm), TiO₂ thickness (0.05–0.4 μm), AZO thickness (0.01–0.10 μm), defect density (10¹²–10¹⁷ cm⁻³), operating temperature (280–360 K), and contact work functions were varied. In the presence of defects, the device achieved an open-circuit voltage (Voc) of 0.859 V, a short-circuit current density (Jsc) of 26.059 mA/cm², a fill factor of 78.42
Perovskite structured LaCoO3 is a promising candidate for rare earth ion doping to improve optical performance because of their broad band gap, high dielectric constant and exceptional chemical stability. This work involved the sol-gel synthesis of Ho³⁺-doped LaCoO₃ (0.1, 0.5, and 1.0 mol
Core-shell nanoparticles with metallic constituents exhibit unique optical properties due to localized surface plasmon resonance (LSPR) associated with them, which enhances the local electric field near the metal nanoparticles. Importantly, LSPR can be tuned across a broad electromagnetic (EM) spectrum by adjusting the core radius and shell thickness, enabling diverse applications in nonlinear optics, surface-enhanced Raman spectroscopy, and biomedicine. This study compares the optical properties of two distinct one-dimensional ternary photonic crystals (1D TPCs) incorporating a nanocomposite (NC) material embedded with core-shell nanoparticles. In the first 1D TPC, the NC layer contains bimetallic core-shell nanoparticles (Ag@Au), while the second 1D TPC features an NC layer embedded with dielectric-core/metallic-shell nanoparticles (SiO2@Au). The optical properties of 1D TPCs were analyzed as a function of core radius, nanoparticle filling fraction within the NC, and angle of incidence. Variations in the metallic core/shell size and nanoparticle filling fraction alter the LSPR of the nanoparticles. These effects result in the deformation of photonic band gap (PBG) edges and the formation of broad absorption features in the spectra of 1D TPCs. The angle of incidence variation tailors the scattering effect in NC, which subsequently affects the LSPR and, in turn, influences the bandgap and absorption spectra of the 1D TPCs. The proposed structures may have potential applications in the design of various optical devices, including solar cells, optical filters, sensors and absorption-based instruments.
In this work, a comprehensive numerical analysis of thin and thick samples exhibiting local and non-local third-order nonlinearity is presented using Gaussian beam decomposition and Fourier transform methods within the Z-scan technique. Analytical formulas for the normalized transmittance are developed for local and non-local nonlinear media. Different values of the non-locality parameter m are investigated for refractive, absorptive, and combined refractive-absorptive contributions. The general theoretical framework is applicable to materials such as ZnO, but the present work focuses on the analytical formalism rather than experimental validation. The derived expressions unify thin and thick medium descriptions, extending previous models to arbitrary non-locality values in thick media. The influence of the non-locality parameter on the Z-scan curve characteristics is systematically analyzed, and the results demonstrate that increasing m reduces the amplitude of the curve, narrows the valley and peak widths, and modifies the asymmetry of the normalized transmittance. Short-pulse laser cavities and optical limiter designs can benefit from the derived analytical expressions for optimizing device performance.
Organic solar cells (OSCs) have achieved significant efficiency improvements in recent years; however, identifying the dominant factors governing power conversion efficiency (PCE) in heterogeneous experimental datasets remains a major challenge. In this work, a physically interpretable machine learning framework is developed to investigate the descriptor hierarchy controlling OSC performance using a curated dataset of 487 devices with 28 electronic, structural, material, and processing parameters collected from literature published between 2020 and 2025. A systematic workflow combining data preprocessing, feature engineering, multiple regression models, and grouped cross-validation is implemented to ensure robust and unbiased predictive modeling. Beyond prediction, the study emphasizes interpretability through correlation analysis, feature importance ranking, SHAP (SHapley Additive exPlanations), partial dependence analysis, and descriptor interaction mapping. The results demonstrate that OSC efficiency is governed by a coupled and nonlinear descriptor space, in which structural and processing parameters particularly active layer thickness exert a dominant influence over purely electronic descriptors. While electronic properties such as HOMO, LUMO, and bandgap define the energetic framework of device operation, their impact on PCE is strongly dependent on morphological and processing conditions. Furthermore, cluster-based and high-performance device analyses reveal that efficient OSCs are confined to a narrow multidimensional design space characterized by coordinated optimization of electronic alignment, morphology, and fabrication parameters. Overall, this study provides a comprehensive and experimentally grounded framework that bridges predictive modeling and physical understanding, offering actionable insights into descriptor interactions and guiding the rational design of high-efficiency organic solar cells.
