Machine learning empowered surface growth of 2D materials: Synthesis, characterization and mechanism

The advent of atomically thin two-dimensional (2D) materials provides a versatile platform to transcend the fundamental limitations of silicon-based electronics and continue the miniaturization of field-effect transistors, yet the epitaxial growth of wafer-scale, single-crystalline structure remains a formidable challenge. In recent years, the vigorous development of machine learning (ML) techniques has contributed to a revolutionary shift in materials synthesis, characterization and application, offering unprecedented opportunities for scientific and technological innovations that are inaccessible through traditional experimental and computational methods. This review aims to outline recent progress of ML-assisted 2D materials growth, including optimizing synthesis conditions, automating real-time characterizations and unveiling growth mechanisms. Current challenges and future prospects in this frontier research field are also discussed. Overall, this review highlights the synergy between advanced ML techniques and surface growth approaches for accelerated materials synthesis and intelligent design of next-generation functional devices.