The Effect of Microstructure and Impurities on Transport in Actinide Oxide Films

The aim of the proposed project is to develop an atom-level description of the actinide oxide films to predict the likely microstructure, the segregation of impurities and their effect on oxygen and hydrogen transport. We will use a combination of DFT and potential model techniques. Details are given in work packages (WP), which define the milestones that this project will achieve.
WP1. Microstructure Generation. The current state of the art is to model surfaces or grain boundaries. The challenge for modelling the properties of the thin films is now to develop models for polycrystalline thin films, where both surfaces and grain boundaries are present. This level of complexity is necessary as these interfaces will provide the pathways for fast diffusion and they can have a profound implication for grain growth. Thus the aim of this WP is to extend the simulation tools so that we can screen through the likely structures and hence use the structures and energies to predict plausible microstructures given different possible operating conditions. These microstructures will be highly dependent on the nature of the defects present and hence we will also investigate how the interfaces and their compositions are affected by intrinsic defects.
WP2. Oxygen and Hydrogen Transport. An understanding of the adsorption and transport of oxygen and hydrogen is central to understanding and ultimately controlling the kinetics of the growth and dissolution processes of the actinide oxides, which in turn will have a profound impact on the material properties. Initially, we propose to use nudged-elastic-band DFT to calculate the activation energies for migration of oxygen in the oxide layer and investigate how it changes depending on the structural features (grain boundaries and thickness of the oxide film). We will investigate the transport through the oxide film by performing constrained molecular dynamics on a range of different microstructures from which we can calculate the migration rates. Once we have evaluated the significant pathways, we will use kinetic Monte Carlo to calculate the overall rates. The latter will allow us to go beyond the nanosecond time scale of direct molecular simulations and to make direct comparison with available experimental data. The behaviour of different structural features within the UO2 matrix will be used to derive quantitative structure property relationships.
WP3. Effect of impurities. All real materials are impure, and even ppm quantities of impurities are often sufficient to coat interfaces and thus the compositions of the interfaces and hence their properties may differ significantly from the bulk materials. Thus we will introduce a range of impurities to provide reliable data on their influence on the diffusion along the boundaries. The influence of doping on the transport properties in fluorite structures with representative alkaline, transition, lanthanide and actinide metals as well as p-elements will be determined. A quantitative relationship will be calculated. Furthermore, we may find that certain impurities show beneficial effects and hence we will be able to provide predictions on which dopants might be added to improve the material properties. The focus will therefore be on doping hyper-stoichiometric UO2 and use a combination of molecular dynamics and kinetic Monte Carlo.