Back to Research
AMDD

Accelerated Materials Design and Discovery

THE CHALLENGE

On the path toward meeting the goals of its Environmental Challenge 2050, Toyota forecasts that its annual global sales of electrified vehicles (HEVs, PHEVs, BEVs, and FCEVs) will grow to nearly 5.5 million units by around 2025. To make this a viable business strategy, batteries must be smaller, lighter and cheaper, last longer, have longer range and charge quicker. Hydrogen fuel cells must meet similar targets.

Significantly improving performance and reducing cost of electrified mobility requires the design and discovery of new materials. Only a handful materials have proven commercially viable for vehicle batteries and fuel cell catalysts; an especially small number when compared to the limitless and largely unexplored number of possible new materials. Materials discovery is inherently a time-consuming process, historically measured in decades. As an example, development of the lithium-ion battery took approximately 40 years to reach viability for vehicle applications. And yet, despite the large R&D investments in lithium-ion over that time, current batteries continue to have deficiencies which limit mass adoption for EVs.

Materials for energy applications have multiple technical requirements, but there are major challenges in trying to optimize several different properties at the same time.

OUR APPROACH

TRI's accelerated materials design and discovery (AMDD) program is working to build the tools and the capabilities for rapidly discovering new materials that are effective for future use in both vehicle batteries and fuel cells - amplifying the abilities of scientists and engineers. Several recent advances in computing power, artificial intelligence and automation plus decreased cost for simulation and data management are making this goal reachable.

Our vision is to develop energy materials that are smaller and lighter, store more energy, last longer, cost less and can be more widely deployed than those currently in use.

OUR WORK
Team

TRI has created a collaborative ecosystem for conducting materials research. We are investing about $35 million in research projects at leading academic institutions. Our consortium has 12 distinct projects spread over 13 research universities and organizations, including the California Institute of Technology, Carnegie Mellon University, Danish Technical University, Georgia Tech, Lawrence Berkeley National Lab, the Massachusetts Institute of Technology, Northwestern University, Stanford University, and the University of Michigan.

Our projects engage approximately 30 principal investigators, 120 graduate students, postdocs, and research staff in addition to 10 dedicated TRI researchers with a blend of expertise from materials science to software engineering. Consortium members regularly communicate and share individual progress and results for furthering the overall effort.

The projects vary in focus ranging from physics-based simulation for prediction of material performance to lab-based experimentation of material combinations. Our research teams are developing novel automated materials discovery systems by integrating simulation, machine learning, artificial intelligence and robotics. The ultimate vision is to operate an automated system of high-throughput computation and experiments that form a cycle of discovery which uses AI to amplify a researcher's ability to discover new materials. The cornerstone of this is extensive, dynamic databases of materials information, which our program is also creating.

OUR PROGRESS
video thumbnail
Video Thumbnail

WATCH VIDEO

TRI's project with Carnegie Mellon University is building a system that can automatically find better electrolytes for batteries. The goal is to discover the best electrolyte-electrode combination that is lower cost, safer, more robust, less toxic and contains more energy with a longer life-expectancy.

video thumbnail
Video Thumbnail

WATCH VIDEO

MIT, Stanford and TRI Researchers discover how to accurately predict the cycle life of lithium-ion batteries using early cycle data and machine learning. Full research findings published in Nature Energy - March 25, 2019.

RESEARCH PAPERS
April 2018

Tabor, Daniel P.; Roch, Loïc M.; Saikin, Semion K.; Kreisbeck, Christoph; Sheberla, Dennis; Montoya, Joseph H.; Dwaraknath, Shyam; Aykol, Muratahan; Ortiz, Carlos; Tribukait, Hermann; Amador-Bedolla, Carlos; Brabec, Christoph J.; Maruyama, Benji; Persson, Kristin A.; Aspuru-Guzik, Alán

Read Paper

March 2019

Severson, Kristen A.; Attia, Peter M.; Jin, Norman; Perkins, Nicholas; Jiang, Benben; Yang, Zi; Chen, Michael H.; Aykol, Muratahan; Herring, Patrick K.; Fraggedakis, Dimitrios; Bazant, Martin Z.; Harris, Stephen J.; Chueh, William C.; Braatz, Richard D.

Read Paper

Oct. 2019

Patel, Anjli M.; Nørskov, Jens K.; Perssonc, Kristin A.; Montoya, Joseph H.

Read Paper

May 2019

Aykol, Muratahan; Hegde, Vinay I.; Hung, Linda; Suram, Santosh; Herring, Patrick; Wolverton, Chris; Hummelshøj, Jens S.

Read Paper

Read Blog Post "Behind the Paper"

Sept. 2019

Ward, Logan; Aykol, Muratahan; Blaiszik, Ben; Foster, Ian; Meredig, Bryce; Saal, James; Suram, Santosh

Read Paper

Nov. 2019

Aykol, Muratahan; Hummelshøj, Jens S.; Anapolsky, Abraham; Storey, Brian D.; Aoyagi, Koutarou; Bazant, Martin Z.; Bligaard, Thomas; Braatz, Richard D.; Broderick, Scott; Cogswell, Daniel; Dagdelen, John; Drisdell, Walter; Garcia, Edwin; Garikipati, Krishna; Gavini, Vikram; Gent, William E.; Giordano, Livia; Gomes, Carla P.; Gomez-Bombarelli, Rafael; Gopal, Chirranjeevi Balaji; Gregoire, John M.; Grossman, Jeffrey C.; Herring, Patrick; Hung, Linda; Jaramillo, Thomas F.; King, Laurie; Kwon, Ha-Kyung; Maekawa, Ryosuke; Minor, Andrew M.; Montoya, Joseph H.; Mueller, Tim; Ophus, Colin; Rajan, Krishna; Ramprasad, Rampi; Rohr, Brian; Schweigert, Daniel; Shao-Horn, Yang; Suga, Yoshinori; Suram, Santosh K.; Viswanathan, Venkatasubramanian; Whitacre, Jay F.; Willard, Adam P.; Wodo, Olga; Wolverton, Chris

Read Paper

Back to Top