A Lab research team’s project was selected in a highly competitive $26 million Department of Energy (DOE) grant program to advance chemical and materials sciences by using data science. The goal of the funding program is to harness cutting-edge research tools for new scientific discoveries fundamental to clean energy solutions.
“This award is a vote of confidence in Los Alamos National Laboratory expertise, as well in the potential of data science to solve critical problems,” said Deputy Laboratory Director John Sarrao. “Modeling the behavior of complex systems is an exciting challenge, and we ’re looking forward to seeing the new chemistries and materials systems that are developed from this project.”
Principal investigator Kipton Barros of Theoretical Division will lead a multidisciplinary team of researchers on a three-year, $2.4 million project in partnership with the California Institute of Technology (Caltech) to use data science — including artificial intelligence and machine learning (AI/ML) — to advance the understanding of chemical and materials systems.
The team will use data science to explore three main areas in chemical and materials systems, including the self-assembly of soft materials (materials like plastic that respond easily to heat and other factors); the chemical processes of catalysis (a substance being used to advance a chemical reaction); and the role of stimuli (such as laser light) and external influences that help drive the outcome of chemical processes. The project will build models and simulations that can be applied in experiments. Ultimately, the resulting predictive AI/ML models, validated by experiments, will accelerate discovery of new chemistries and materials systems with exceptional properties and functionalities. The impact of the research may be seen in everything from better materials for solar cells to improved catalysts for chemistry to drug discovery.
Artificial neural net models
A guiding principle of data science used in the project is the development of artificial neural net (ANN) models. Inspired by the concept of neurons in the brain, a neural net is a computer model that can be trained to find patterns in a dataset and that can make new predictions. The partners at Caltech will develop state-of-the-art computational quantum chemistry methods needed for the project. Building on that work, the Los Alamos team will develop ANN models to extend the existing modeling capabilities and to achieve large length and timescale dynamical simulations necessary for understanding chemical processes.
“In chemistry, to get a product you want, you may need multiple steps,” said Barros. “It can be hard to uncover the most efficient sequence of steps. That’s where the data science can help us. The big picture of our project is for the computer to give us that insight that would be extremely difficult to get otherwise.”
The project began Sept. 1 and will run for three years. More information about the DOE Office of Science’s Basic Energy Science programs, including the Data Science to Advance Chemical and Materials Sciences funding opportunity, can be found here.