UC Davis

Uranium-based materials are an important research topic in material science due to numerous industrial and scientific applications. However, hydrogen embrittlement of uranium, which arises due to the formation of a structurally weak pyrophoric hydride, poses a major safety risk in material applications. Relatively little is known about the hydriding initiation mechanism in pure uranium, in part due to the material’s highly reactive nature and toxicity. Future hydriding studies would thus greatly benefit from atomic-level simulations of the hydriding process, which can provide microscopic details about the hydrogen-uranium reaction and help guide future experimentation. Kohn-Sham Density Functional Theory (DFT) is a well-established quantum mechanical technique that can be used to elucidate many aspects of the onset of hydride formation. DFT is used in chemistry, physics, and materials science for accurate prediction of physical and chemical properties, such as material equations of state, heats of formation, and the energetics of bond forming/breaking under reactive conditions. However, molecular dynamics (MD) simulations run with DFT are generally limited to time scales on the order of picoseconds and system sizes of a few hundred atoms. In contrast, many processes related to hydriding, such as initiation, nucleation, and growth, probe significantly longer time and length scales, and can require simulation cells of thousands of atoms run for nanosecond timescales or longer. In this regard, the Chebyshev Interaction Model for Efficient Simulation (ChIMES) approach is a method for rapid creation of reactive MD models. Briefly, ChIMES is optimized through determination of linear combinations many-body Chebyshev polynomials by fitting to DFT simulation data, such atomic forces, system energies, and stress tensor components. ChIMES has been shown to yield similar accuracy to DFT while yielding linear scaling and orders of magnitude improvement in computational efficiency. In this dissertation, we first present a systematic investigation of possible mechanisms for the formation of the metal hydride using DFT. Specifically, we address this problem by examining the individual steps of hydrogen embrittlement, including surface adsorption, subsurface absorption, and the interlayer diffusion of atomic hydrogen. Furthermore, by examining these processes across different facets, we highlight the importance of both (1) hydrogen monolayer coverage and (2) applied tensile strain on hydriding kinetics. Taken together, by studying these previously overlooked phenomena, our study provides foundational insights into the initial steps of this overall complex process. Next, we use DFT generated data to develop a ChIMES U-H model that is fit to a training set containing energies and forces of U and UH3 bulk structures with vacancies and hydrogen interstitials. We show that the bulk structural parameters, point-defect formation energies, and diffusion barriers predicted by the ChIMES potential are in strong agreement with the reference DFT data. We then use ChIMES to conduct MD simulations of the temperature-dependent diffusion of a hydrogen interstitial and determine the corresponding diffusion activation energy. Our work is likely to have impact in research areas where there traditionally is a strong need for computationally efficient methods to bridge length and time scales between experiments and quantum theories.