Mohammad Istiaque Hossain

Infrared (IR)-induced heating in photovoltaic (PV) systems is a critical challenge that lowers efficiency and accelerates module degradation. In this work, we propose a multilayer thin-film IR filter (TiO2 (50 nm)/NiOx (100 nm)/Ag (various thickness)) as a reduced heating solution integrated with PV modules. The filter is designed to transmit visible light while reflecting IR radiation, thereby reducing thermal load without sacrificing photovoltaic current. We incorporate machine learning models specifically Gaussian Process Regression (GPR) to optimize the Ag layer thickness for maximum performance. The AI models are trained on a combination of experimental optical data, enabling efficient exploration of the thickness-performance space. Key findings demonstrate that an optimized Ag thickness (∼10 nm) yields high IR reflectance (over 50% in the 750–1200 nm range) while maintaining sufficient visible transmittance (>50%). Silicon solar cells with this filter showed improved performance: the short-circuit current and power output increased due to reduced thermalization losses, translating to a ∼2–3 °C drop in operating temperature and corresponding efficiency gains. These improvements can extend PV module lifespan and energy yield. Our results indicate that the TiO2/NiOx/Ag filter can be manufactured via scalable e-beam evaporation, and the integration of AI optimization accelerates the design of such photonic coatings. The demonstrated performance enhancements and the low-cost, scalable nature of the solution highlight strong potential for commercial deployment in extending PV module longevity. We adopt Gaussian Process Regression (GPR) as a surrogate tailored to small, high-fidelity experimental data sets. GPR provides calibrated predictive uncertainty and smooth, physics-consistent interpolation across film thickness and wavelength, enabling uncertainty-aware selection of robust thicknesses with few experiments. This contrasts with polynomial/linear fits (insufficiently expressive) and black-box models without calibrated uncertainty.

The extreme conditions of deserts such as intense solar irradiation and elevated dust concentration in the air pose significant operational issues to photovoltaic (PV) modules, optical components, and exposed functional surfaces. This research paper has presented metal-integrated silica thin-film coating materials that can be used to address dust adhesion, optical reflection loss, and environmental degradation at the same time. In this study, we investigate the metal integrated silica coatings with controlled optical and morphological properties using RF magnetron sputtering. 100 nm-SiO₂, 10 nm-Ti/100 nm-SiO₂, and 10 nm-ITO/100 nm-SiO₂ were developed and examined systematically to understand the dust adhesion mechanism for two months. The Ti/SiO₂ optimized coatings showed lower dust adhesion, and this is explained by the combined effect of the morphology, wettability, and surface energy. The potential dissipation of the electrostatic charge is expected as one of the factors. In addition, surface morphology strongly depends on the types of materials deposited on silica layers as found by average roughness (R_a). Contact angle measurements confirm that the hydrophilicity of such engineered coatings can be tuned through metal content in the films. Micromechanical properties of the coatings were thoroughly studied immediately after sputtering deposition and after 2 months of outdoor exposure using Vickers microindentation analysis. Among all, as-deposited Ti/SiO₂ coatings show excellent mechanical durability after a period of 2 months. As studied computationally, the power conversion efficiency (PCE) is predicted to reach almost 27.28% in the case of 10 nm Ti/100 nm silica and decreases to less than 8% at 20 μm. Ti/silica layers of the device play a crucial role in device performance. Our investigation also found that the adhesion of the dust to the films reduces as the metal films are integrated into the silica layers, which confirms the intrinsic self-cleaning nature of the coatings. Our works also investigated that dust adhesion on the surface of these metal-integrated silica coatings is inversely proportional to the metal contents in the films, confirming the coatings’ self-cleaning properties. The hydrophilic characteristics of the sputtered thin films can limit the adhesion of particles to the surface, which is crucial to enhance the performance of photovoltaic systems in desert areas. Such a solution can enhance both optical and mechanical characteristics as well as the performance of PVs in harsh conditions.

Thin film thermoelectrics offer compact energy-harvesting and precise cooling solutions for micro-electronics central to Internet of Things and Artificial Intelligence technologies. Advancing these applications requires rapid and reliable experimental screening methods to identify high‑performance thermoelectric materials at the thin‑film scale. Here, we present a semi‑automated, Python‑controlled platform for fast room‑temperature evaluation of thin‑film thermoelectric power factors. The system integrates a 3D‑printed sample holder with built‑in thermocouple probes, a sourcemeter, and a multimeter, all operated through a custom graphical user interface that enables streamlined, repeatable measurements of electrical conductivity and Seebeck coefficient with minimal user input. Benchmarking against metallic (Au, Ag, Ti) and transparent conducting oxide (ITO, FTO) thin films demonstrates excellent agreement with literature values, accurately capturing both the magnitude and sign of the Seebeck coefficient across substrates with varying thermal conductivities. The metallic films exhibit small, positive Seebeck coefficients consistent with complex electronic structures near the Fermi level, while the transparent oxides show small negative values characteristic of degenerately doped n‑type semiconductors. This robust, high‑resolution platform provides a practical tool for rapid laboratory‑scale screening and optimization of thin‑film thermoelectric materials.

Thin metal films are essential for expanding sensors, optoelectronic, and photovoltaic technologies. The intricate relationship between material composition, thickness, and production presents significant challenges in optimizing optical properties. The paper introduces an AI-driven framework for the simultaneous prediction and optimization of metal film optical characteristics, such as transmittance(%T), reflectance(%R), and absorptance(%A) using a dataset of 1320 experimentally measured samples across material films of gold, aluminum, nickel, tin, copper, and molybdenum over 200–2000 nm wavelength range. The input features to the model include wavelength and material type. Ensemble Models such as Random Forest, Gradient Boosting, XGBoost, and Extra Trees were trained and optimized through GridSearchCV with stratified K-fold cross-validation. A multitask learning model was also implemented to explore potential improvements from joint prediction. Among all models, the CatBoost Regressor demonstrated superior performance, achieving R ² = 0.99928, MAE = 0.21924, and MSE = 0.28203 on average across all outputs. To enhance interpretability, the feature importance analysis was employed, revealing that Material Type had a slightly more predictive influence than Wavelength. Additionally, material-specific error analysis identified challenging prediction zones tied to spectral extremes. The best performing machine learning model was deployed via a web-based GUI, enabling real-time prediction of thin-film optical properties. Overall, the proposed framework provides a scalable, interpretable, and deployable solution for AI-assisted material design and optical characterization. These findings accelerate thin metal film optimization by offering a reliable data-driven route for quick material property identification and enhancement through machine learning.