Reprogramming Photocatalysis Through Self-Evolving Interfaces Toward Adaptive Reaction Systems

Photocatalysis has traditionally advanced through material-centric strategies, including band engineering, heterojunction construction, and defect modulation, aimed at enhancing light absorption and charge separation. However, improved charge separation does not necessarily translate into efficient or selective chemical conversion, particularly when catalytic interfaces evolve under operating conditions. Here, we review photocatalysis through the framework of self-evolving interfaces, in which electronic structures, surface states, active sites, and adsorbate environments change in response to illumination and reaction conditions, and we classify this evolution into reversible responses, productive reconstruction, deactivation, and feedback-controlled adaptation using experimentally testable criteria. This framework shifts the focus from carrier generation and separation toward reaction-oriented charge utilization, in which charge localization, interfacial transfer, and intermediate stabilization determine reaction pathways and product selectivity. We extend this perspective to the reactor and system levels by considering the coupling of photon transport, mass transfer, and reactor configuration, as well as hybrid integration with photovoltaic, energy-storage, and electrochemical components for managing intermittent solar input. Finally, we discuss how operando characterization and data-driven analysis can identify productive interfacial states and guide feedback-based operation, and we outline the remaining challenges toward efficient, selective, and ultimately adaptive solar-to-chemical systems.