Visual analysis of defect clustering in 3D irradiation damage simulation data

Molecular dynamics simulation has become a powerful tool to deepen the understanding of the radiation damage mechanism of nuclear materials. Extracting point defects, analyzing their diffusion, and visualizing the defect dynamics in atomic simulation data are important, but challenging tasks to understand irradiation behavior. In the past, irradiation defects have been detected using the so-called Wigner–Seitz cell method and analyzed by the statistics of Frenkel pairs. However, traditional analysis modes blur the fine details of defect dynamics. In this paper, we present a visual analysis pipeline for domain scientists to comfortably explore radiation damage simulation data. We couple defect identification, defect clustering, molecule visualization, and tracking graph to form an integrated visual exploration approach. We describe the application of our approach in practice to study defect clustering in Ni–Fe alloy. With our proposed pipeline, defects can be extracted in a robust way, clusters can be visualized with a favorable representation, in-depth data analysis can be setup, and defect dynamics can be demonstrated in greater detail than previously possible.