Graphene membranes can improve the performance of reverse osmosis water desalination systems if the nanopore geometry of the 2D material is optimized. The excessive computational, time and experimental cost incurred in determining optimal nanopore configurations prompted Carnegie Mellon University researchers to develop an artificial intelligence (AI)-based method to streamline membrane design.

The computational framework combines a deep reinforcement learning algorithm with a convolutional neural network performance predictor to discover the optimal graphene nanopore for water desalination. The method detailed in the journal npj 2D Materials and Applications was demonstrated to rapidly create and screen thousands of graphene nanopores and select the most energy-efficient ones.

Molecular dynamics simulations on promising AI-created graphene nanopores show that they have higher water flux while maintaining rival ion rejection rate compared to the normal circular nanopores. Irregular shape with rough edges geometry of AI-generated pores was observed to be a key factor for enhanced water desalination performance.

The research confirms that AI can be a powerful tool for nanomaterial design and screening. The approach developed will help users determine the best membranes for desalination and separation, expanding its use across multiple disciplines.

To contact the author of this article, email