Key points are not available for this paper at this time.
Abstract Low‐dimensional materials have attracted significant interest for their unique properties, including high surface area, confined but tunable electronics and superior catalysis, making them ideal for environmental applications. Their potential to address key challenges in solar energy conversion and in‐situ remediation highlights their importance in advancing environmental sustainability. However, traditional methods of low‐dimensional material design face significant obstacles, such as scalability limitations, high computational costs, and the inherent difficulty in accurate prediction of material properties, underscoring the need for innovative approaches. Here, we demonstrate an AI‐driven evolution of low‐dimensional material design for sustainable environmental solutions, from the traditional techniques in the past, through the present transition to computational approaches, to the prospect where AI‐enabled strategies exhibit the supremacy. This review covers properties of low‐dimensional materials and the fundamental design principles, emphasizing the pivotal role of deep learning in optimizing and accelerating design of advanced functional materials. Further explorations focus on their applications for sustainable environmental solutions, including pollution remediation, water purification, nitrogen fixation, CO 2 reduction as well as hydrogen and hydrogen peroxide production. Ultimately, the key challenges and future trends are identified in the aspects of algorithm, intelligence and scalability for environmental applications. This work offers a comprehensive overview on the evolution pathway of design strategies for low‐dimensional materials driven by AI methodology, demonstrating transformative insights that not only accelerate the discovery of low‐dimensional materials, but also motivate the environmental applications in various domains.
Su et al. (Mon,) studied this question.