Recent advances in rational design of defect-engineered photocatalysts toward sustainable NH3 synthesis as H2 carrier: From fundamental and development to machine-learning

The Haber–Bosch process remains a pivotal technology in the synthesis of ammonia (NH 3 ). It has fundamentally transformed agriculture and the chemical industry; however, this method accounts for approximately 1 % of global annual energy consumption for nitrogen-based fertilizer production. This results in significant energy demands and substantial carbon dioxide emissions. As global efforts intensify to mitigate environmental impact, photocatalytic NH 3 synthesis has emerged as a promising, sustainable alternative. It produces nitrogen-containing compounds with lower energy requirements. This approach leverages sunlight as a renewable energy source and water as a reducing agent. Despite its potential, the efficiency of photocatalytic NH 3 synthesis is currently limited mainly due to the scarcity of highly active catalytic sites. Recent research has shown that engineering surface defects on photocatalysts can improve their performance. These defects enhance light absorption, weaken the N N bond of adsorbed nitrogen (N 2 ), promote charge separation, and facilitate charge transfer to adsorbed reactants. Nevertheless, achieving a comprehensive understanding of how modifications to the catalyst's microstructure influence N 2 binding, activation, and hydrogenation remains a significant challenge. These knowledge gaps hinder the development of photocatalysts suitable for industrial use. In this review, we provide a detailed overview of the fundamental mechanisms underpinning photocatalytic N 2 reduction. We also discuss advances in catalyst design for the synthesis of NH 3 . Particular emphasis is placed on the role of surface defect engineering, which includes the creation of surface defects to enhance the performance of semiconducting photocatalysts for efficient N 2 reduction. In addition, the application of a machine learning-based computational modeling approach is discussed as an important driving force for predicting and regulating NH 3 synthesis efficiency based on catalyst features and reaction conditions. Finally, existing challenges and future perspectives for improving the performance of defect-engineered photocatalysts are outlined to contribute to the ongoing discourse on sustainable ammonia generation. This review aims to clarify recent progress in the rational design of defect-containing photocatalysts for the synthesis of NH 3 and encourages innovative approaches to catalyst optimization rather than solely focusing on new materials. • The fundamental rules and existing challenges in photocatalytic NH 3 were discussed. • A defect engineering strategy for developing advanced photocatalysts for NH 3 synthesis was explored. • Recent progress of anion defects in promoting the photocatalytic NH 3 synthesis were summarized. • Key potentials and outlooks for applications of Machine-learning in NH 3 photo-production are proposed.