Speaker
Ju Li has held faculty positions at the Ohio State University, the University of Pennsylvania, and is presently a chaired professor at MIT. His group investigates the mechanical, electrochemical and transport behaviors of materials as well as novel means of energy storage and conversion. Ju is a recipient of the 2005 Presidential Early Career Award for Scientists and Engineers, the 2006 Materials Research Society Outstanding Young Investigator Award, and the TR35 award from Technological Review. He was elected Fellow of the American Physical Society in 2014, a Fellow of the Materials Research Society in 2017 and a Fellow of AAAS in 2020. Li is the chief organizer of MIT A+B Applied Energy Symposia that aim to develop solutions to global climate change challenges with “A-Action before 2040” and “B-Beyond 2040” technologies.
Date
Friday June 29th, 10am (USA/Pacific)
Abstract: A Universal Empirical Interatomic Potential
Ju will describe the recent invention of a robust universal machine learning interatomic potential that covers much of the periodic table. More than one thousand GPU years were used to generate the ab initio training data guided by active learning. Diverse test simulations have shown this machine learning potential has outstanding performance, with energy error significantly less than the chemical accuracy (1kcal/mol) for even chemically very complex systems. This universal potential can run over a million times faster than density functional theory (DFT) when dealing with several thousand atoms, and the latest release allows for more than 50,000 atoms of arbitrary combinations of 72 elements to be simulated together. One can use this empirical potential to study realistic microstructures such as extended defects with curvatures and their interactions, realistic phase transformations, plastic deformation and damage evolution, electrochemical interfaces, etc. [J. Materiomics 9 (2023) 447]