Evaluation of the Performance of Natural Dye-Sensitized Solar Cells Using Red Spinach and Henna: Combining Fabrication Insights with ML-Based Efficiency Predictions

This research evaluates the effectiveness of various TiO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> paste compositions in natural dye-sensitized solar cells (DSSCs) using machine learning (ML) models to predict their performance under both indoor and outdoor conditions. The study investigates the impact of key factors, including electrolyte layer thickness, use of industrial-grade $\mathbf{T i O}_{2}$, and electrolyte concentration, on cell electrical properties and durability. Natural dyes used for sensitization were red spinach (Amaranthus cruentus) and henna (Lawsonia inermis). Multiple ML models were employed to analyze the influence of these parameters on solar irradiance. Under controlled laboratory conditions, the ML models accurately predicted the performance of DSSCs across various environmental conditions. In comparative tests under the AM 1.5 G solar spectrum, typical of Bangladesh, it can be confirmed that DSSCs sensitized with red spinach achieved a significantly higher efficiency of $\mathbf{0. 8 1 \%}$ compared to those sensitized with henna ($\mathbf{0. 0 1 8 \%}$).