Polarization tuning of ferroelectric photocatalysts: From conventional selection to AI-guided design for efficient CO2-to-C1 conversion

Ferroelectric materials present a transformative platform for the photocatalytic reduction of CO 2 , primarily due to their tunable internal polarization. This review focuses on ferroelectric polarization, elucidating polarization switching mechanisms and polarization effects induced by external fields. We analyze asymmetric transitions and dipole rotation within domain structures, emphasizing intrinsic material properties and field-driven domain wall dynamics. We systematically examine CO 2 adsorption/activation mechanisms on ferroelectric surfaces and correlate reaction pathways with product selectivity. Critically, we dedicate a final section to the emerging paradigm of AI-guided design, discussing how machine learning models accelerate the discovery of photocatalysts and predict optimal configurations for maximized C 1 product yield. Further, we explore advanced design strategies to optimize polarization and enhance CO 2 reduction efficiency via morphology engineering, defect/atomic engineering, strain engineering, and interface engineering. By synthesizing current experimental and theoretical foundations, this review aims to provide a comprehensive roadmap for harnessing ferroelectricity to advance renewable energy technologies.