Models and regressions to describe primary damage in silicon carbide.

Silicon carbide (SiC) and SiC/SiC composites are important candidate materials for use in the nuclear industry. Coarse grain models are the only tools capable of modelling defect accumulation under different irradiation conditions at a realistic time and length scale. The core of any such model is the so-called "source term", which is described by the primary damage. In the present work, classical molecular dynamics (MD), binary collision approximation (BCA) and NRT model are applied to describe collision cascades in 3C-SiC with primary knock-on atom (PKA) energy in the range 1-100 keV. As such, BCA and NRT are benchmarked against MD. Particular care was taken to account for electronic stopping and the use of a threshold displacement energy consistent with density functional theory and experiment. Models and regressions are developed to characterize the primary damage in terms of number of stable Frenkel pairs and their cluster size distribution, anti-sites, and defect type. As such, an accurate cascade database is developed with simple descriptors. One of the main results shows that the defect cluster size distribution follows the geometric distribution rather than a power law.