Researchers from the University of Illinois Urbana-Champaign developed a method to simulate diffusion in multicomponent alloys using “kinosons” and machine learning. This approach breaks down diffusion into individual contributions, significantly speeding up calculations by 100 times compared to traditional methods. The method provides detailed insights into atomic movements and diffusion processes in materials like steel and batteries, enhancing our understanding and potentially improving material design.
For more details, read the full article here: