Total: 1
Binary collision approximation (BCA) codes are potentially powerful tools to simulate ion irradiation ejecta properties, such as the composition and the angular and energy distributions of the sputter yield. However, recent advances in the sputtering of minerals have highlighted the low predictive fidelity of BCA codes such as SDTrimSP when compared to experimental measurements. We demonstrate how a sputtering model that underestimates the forward sputtering on a flat surface at large ion incidence angles from surface normal will lead to an erroneous result for rough and porous surfaces, where most ejected particles are directed along the surface normal. We demonstrate how this is the case for an existing model, which reliably predicts sputtering mass yields from a flat enstatite surface but fails to accurately reproduce the angular distribution of sputtered particles. We then compare this to a BCA model incorporating higher surface-binding energies$-$based on a molecular dynamics description of plagioclase$-$which underestimates mass yields but significantly reduces back-sputtering and better reproduces laboratory sputter angle distributions measured at large ion incidence angles. We conclude that the BCA model cannot simultaneously reproduce both the sputter yield and the sputter angle distribution arising from He irradiation of mineral targets, either due to the inherent geometric simplicity of the BCA or because the model neglects yield-enhancing processes such as molecule and cluster sputtering. This demonstrates a structural limitation of current BCA-based models when realistic surface morphologies are considered, rather than a problem that can be resolved by parameter tuning alone.