The U.S. Department of Energy’s Ames National Laboratory will launch a new research effort, the ML-Accelerated Materials Discovery Center, to drive the discovery and design of complex magnetic and superconducting materials for energy and technology applications.
Materials discovery and design is often a first and fundamental step in developing a new technology. The materials that have just the right combination of unique properties to make a new technology possible are often complex, made up of three or more ingredients with multiple ways to construct them. They exist in nearly endless permutations and combinations to consider in order to find a recipe for technology success—a scientific needle in a haystack. Computer codes developed by Ames researchers show effective methods of searching for structures, and of utilizing machine learning to accelerate the search.
Based on this success, the new project seeks to develop the capability to use some of the most powerful computers in the world, called exascale supercomputers, to find the right choice from extremely large and complex sets of possible solutions. Paired with Artificial Intelligence and Machine Learning (AI/ML) tools, computational scientists at Ames Lab seek to utilize exascale computing to speed the discovery of new materials, and in doing so, speed the discovery and application of new technologies.
“This project is the next logical advancement of Ames National Laboratory’s expertise in using computational theory and design to speed the discovery of new materials for energy and other applications,” said Ames National Laboratory computational physicist and project leader Cai-Zhuang Wang.
The challenge is harnessing the many processors and multiple types of hardware present on the most powerful machines. To do this, Ames Lab’s new center will utilize software developed through the Exascale Computing Project (ECP), a joint project between the U.S. Department of Energy’s Office of Science and the National Nuclear Security Administration which laid the groundwork for a national exascale computing ecosystem. The new computational codes and workflows will integrate materials theories, methods, and databases with AI/ML tools. The center will access some of the nation’s most powerful computers, located at other national laboratories, to demonstrate the readiness of their software, and to discover and design new compounds as a proof of principle.
The goal is to dramatically speed up the prediction of potential new materials and to identify possible pathways for their synthesis.
“Optimizing computational tools to accelerate the identification of new materials and their synthesis pathways empowers experimental researchers to focus on the solutions that are most likely to spell success for a developing technology,” said Wang.
The ML-Accelerated Materials Discovery Center will be operated in collaboration with Los Alamos National Laboratory, and is funded by the U.S. Department of Energy Office of Science. The center supports the DOE’s missions in Exascale Computing and Artificial Intelligence for Science, and aligns with the goals of the Materials Genome Initiative (MGI).
Ames National Laboratory is a U.S. Department of Energy Office of Science National Laboratory operated by Iowa State University. Ames Laboratory creates innovative materials, technologies, and energy solutions. We use our expertise, unique capabilities, and interdisciplinary collaborations to solve global problems.
Ames Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, please visit https://energy.gov/science.