Designing multilayer graphene membranes with well seawater desalination performance using machine learning combined with multi-objective optimization

Given the water scarcity problem, the development of seawater desalination technology with high efficiency and cost control is important. With outstanding separation efficiency, graphene-based membranes have already become a potential choice for this problem. In this work, we employed a combination of Machine Learning (ML) and the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to assess the impact of various parameters on the desalination performance of multilayer graphene membrane systems. The machine learning result indicates that gap width is a critical factor influencing water flux, while layer spacing and offset significantly affect ions rejection rate. Molecular Dynamics (MD) simulations confirm that multilayer graphene membrane ( dG = 1.35 nm, H = 0.85 nm, O = 1.96 nm, and Q = −0.09 e/atom) identified through the NSGA-II optimization process exhibit exceptional desalination performance, characterized by high water flux and increased ions rejection rate. Furthermore, the mechanistic analysis revealed that Na + ions require to shed a greater number of water molecules when passing through the multilayer graphene membranes with smaller layer spacing, resulting in increased energy loss . The integration of ML and NSGA-II can facilitate the design of materials with excellent desalination performance. Designing multilayer graphene desalination membranes based on the combined machine learning and multi-objective optimization algorithm. The specific impacts of various factors on the desalination performance of multilayer graphene membranes were analyzed. • The desalination performance prediction models of multilayer graphene membranes are established. • The multi-objective optimization design of desalination performance based on machine learning and NSGA-II algorithm is carried out. • Gap width is a critical factor influencing water flux, while layer spacing and offset significantly affect ions rejection rate. • The factors influencing desalination performance are analyzed.