Polynomial-Optimized Calcination of TiO₂-Based Photoanodes for Enhanced Photocurrent in Solar Hydrogen Production: Integrating Data Science with Functional Design

The calcination temperature of metal oxide semiconductors plays a pivotal role in dictating their crystallinity, phase purity, and charge transport behavior, all of which are critical to enhancing photoelectrochemical (PEC) water splitting. In this study, we apply a fifth-order polynomial curve fitting model to analyze and optimize the calcination temperature of various n-type semiconductors, ultimately identifying 550 °C as the first-layer calcination temperature for titanium dioxide (TiO₂) photoanodes. Based on this data-driven insight, bimetallic oxide composites of ZnO/TiO₂ (1:2) and MnO₂/TiO₂ (1:2) were synthesized via solid-state mixing followed by thermal treatment at the polynomial-predicted temperature. X-ray diffraction (XRD) analyses confirmed the formation of tetragonal anatase TiO₂, hexagonal wurtzite ZnO, and β-MnO₂ crystalline phases in their respective composites. Structural refinement revealed an increase in crystallite size and d-spacing shifts, indicative of heterojunction formation. Photoelectrochemical measurements under simulated sunlight revealed that MnO₂/TiO₂ (1:2) exhibited the highest photocurrent density (0.02052 A cm⁻²), outperforming pristine TiO₂ and ZnO/TiO₂ (1:2) composites. Electrochemical impedance spectroscopy (EIS) further demonstrated a reduced charge transfer resistance in MnO₂/TiO₂ (9.76 Ω), supporting its superior PEC performance. This work demonstrates a successful integration of predictive modelling with materials synthesis, offering a rational and reproducible pathway for optimizing thermally treated photoanodes for solar-driven hydrogen generation