Material property predictions from simulations have reached an incredible inflection point of capability that can transform your business or research lab.
Understand what limits your system or material performance
Improve system performance by studying system and material parameters
Explore new material performance and interactions through accurate atomic simulations before running costly experiments
How?
Accurate simulations based on first-principles electronic density functional theory can predict reactive properties of materials but are limited to 100s of atoms and picoseconds of time
With machine learning on that data we can extend the reach to time and length scales useful to industry (>1000s of atoms and >nanoseconds of time)
Foundation and custom trained machine learned interatomic potentials unlock these longer scales to predict new dynamic properties directly from simulation
Experience
Complex reactive environments with catalysts
Uncovered proton dynamics in fuel cells at the catalyst layer to enable performance and cost improvements
Designed physically informed machine learned interatomic potentials with on-the-fly learning methods and conformation space exploration and evaluation
Solids and liquids
Predicted transition temperatures of materials using thermodynamic integration and machine learned interatomic potentials
Highly inhomogeneous liquids
Designed custom methods to train solvation models for classical density functional theory
We use cookies to analyze website traffic and optimize your website experience. By accepting our use of cookies, your data will be aggregated with all other user data.