(86d) Chemistry Under Extreme Conditions: Accelerating Research through Artificial Intelligence

Many of the most challenging problems in chemical engineering involve reactivity in complicated systems. First principles methods like density functional theory (DFT) and density functional tight binding (DFTB) are attractive for studying these problems due to their predictive power, however, both methods are confined to systems of a few 100s of atoms, and in the case of DFT, 10s of ps. These limitations have motivated extensive work towards development of accurate, scalable, and reliable reactive interatomic potentials but these models generally rely on complicated functional forms requiring time consuming parameterization schemes. As a result, the physiochemical space over which parameters are available is limited.

We have developed ChIMES, an artificial intelligence-based reactive interatomic potential generation and simulation approach, to overcome these challenges. ChIMES is particularly well suited for studies that would otherwise rely on costly quantum simulations, and to date, has been successfully applied to problems spanning chemical reactivity under extreme conditions to hydriding in solid state materials (i.e., problems with characteristic time and lengths many orders of magnitude greater than is accessible to standard DFT). Ultimately, ChIMES has enabled quantum accurate atomistic simulations on scales overlapping with experiments, in many cases, for the first time. Details of the ChIMES framework will be presented along with its application towards achieved an improved understanding of shockwave-driven chemistry in organic molecular materials.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.