Answers are grounded in this paper and cite their sources.
Sputtering of minerals contributes to exosphere formation on airless planetary bodies, yet experiments provide limited information on the composition of sputtered material, and binary collision approximation (BCA) models depend strongly on accurate surface binding energies (SBEs). Molecular dynamics (MD) simulations include many-body interactions and can resolve ejecta characteristics. Here, we use MD to study the sputtering behavior of wollastonite, a common planetary surface analogue, under 2 keV Ar bombardment, comparing the empirical BMP potential and the machine-learning interatomic potential NeuroEvolution Potential against experiments and BCA models. Both potentials accurately capture the experimental sputtering yield. BCA calculations using the SDTrimSP input parameters from SpuBase agree closely with both the MD yields and experiment, with a large deviation above 60° from the surface normal, whereas SDTrimSP using MD-derived SBEs underpredicts the yield. The MD sputtered composition shows preferential sputtering of oxygen, corresponding to ∼65%–72% of the total yield. Calcium is sputtered more readily as single atoms, with silicon sputtered mainly as part of multiatomic clusters. While both potentials closely capture the total yield, their main differences occur during atomic and cluster ejection of the individual constituents, up to 20% for atomic oxygen and up to 40% for oxygen sputtered as part of clusters.
No takes yet. Share an insight, caveat, or question.
Georgiou et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: