Researchers at the University of Notre Dame have used machine learning to identify new polymer membranes that significantly improve the efficiency of heat-free gas separation. By employing graph neural networks, the team discovered materials that can separate gases up to 6.7 times more effectively than previous membranes. This approach, validated through synthesis and testing, revealed that materials typically used in electronics could excel in gas separation, potentially reducing energy consumption and carbon emissions in industrial processes.
For further details, visit the article on Phys.org: