Development of a Model for Irradiation-Assisted Grain Growth for Nanocrystalline UO2

In this work, we have developed a model for irradiation-assisted grain growth in nanocrystalline UO$_2$ using the MARMOT code. We include the impact of irradiation on UO$_2$ grain growth by coupling a phase field grain growth model with a heat conduction simulation that features a random heat source representing thermal spikes. Our model parameters have been calibrated against experimental measurements at 300 K. The calibrated model predicts grain growth in an irradiated UO$_2$ thin film that compares well with experimental data at 50 K. These results suggest that thermal spikes are the major cause of the irradiation-assisted grain growth observed in the UO$_2$ experiments. They also indicate that irradiation-assisted grain growth is only significant with average grain sizes less than 35 nm, and thus can be neglected when considering fuel performance of typical UO$_2$ fuel pellets.

💡 Research Summary

This paper presents the development and validation of a mesoscale model for irradiation-assisted grain growth in nanocrystalline uranium dioxide (UO2). The primary objective was to understand and predict the phenomenon where grain growth occurs under irradiation at cryogenic to room temperatures, a process not driven by thermal activation alone.

The model is implemented within the MARMOT code, built on the MOOSE multiphysics framework. Its innovation lies in the tight coupling of two components: a phase-field grain growth model and a heat conduction model. The phase-field model, based on the work of Moelans et al., describes the curvature-driven evolution of grain boundaries (GBs) using order parameters, incorporating known average GB energy and mobility for UO2. The key to simulating irradiation effects is the incorporation of thermal spikes—localized, transient high-temperature events caused by collision cascades during ion irradiation. This is achieved by introducing a stochastic heat source term (Q) in the heat equation. The heat source activates randomly in space and time, with its magnitude, spatial extent (~4.84 nm radius), and energy (25.65 keV per spike) derived from experimental data and SRIM simulations.

The model parameters were calibrated against in-situ ion irradiation experimental data for nanocrystalline UO2 thin films at 300 K, reported by Yu et al. (2012). Through this calibration, the previously unknown thermal spike duration was estimated to be on the order of picoseconds. The calibrated model was then successfully tested by predicting grain growth under irradiation at a much lower temperature of 50 K, showing excellent agreement with the independent experimental dataset.

The simulation results lead to two major conclusions. First, they strongly suggest that thermal spikes are the dominant mechanism responsible for the observed irradiation-assisted grain growth at low temperatures. The local, extreme temperature increase within a spike momentarily drastically enhances GB mobility, driving boundary migration. Second, the effect exhibits a strong size dependence. The model predicts that irradiation-assisted grain growth is only significant for materials with an average grain size below approximately 35 nm. As grain size increases, the probability of a thermal spike intersecting a GB decreases, diminishing the effect.

Therefore, the study concludes that while irradiation-assisted grain growth is a critical consideration for nanocrystalline UO2 (e.g., in advanced fuel forms or thin films), its impact can likely be neglected for the performance analysis of conventional UO2 fuel pellets, which typically have micrometer-sized grains. This work provides a valuable predictive tool for microstructural evolution in nuclear fuels under irradiation, particularly in regimes where nanoscale features are present.