Molecular Dynamics Simulations of Ion Irradiation of a Surface under an Electric Field S. Parviainen, F. Djurabekova

Ion irradiation during breakdown Breakdown: plasma above surface Ions from the plasma are accelerated towards the surface Irradiation Plasma sheath, strong electric field Irradiation in general is well studied, but what about irradiation under breakdown conditions? Plasma sputtering invented in 1852, but still little data?

What I will tell you Sputtering yield (=sputtered atoms/incident ions) is modified by: The presence of an electric field Surface features (e.g. protrusions) Also modifies local flux Relevance for breakdown research: plasma is maintained by sputtered atoms Input for plasma simulations So far these have not included the modified yields

The modeled system Flat surface, flat surface w/ protrusion Cu+ ions on Cu surface perpendicular to surface DC - constant electric field Room temperature “Low” flux (No multiple hits in same spot) Hybrid Molecular Dynamics - Electrodynamics See Flyura's talk for details

Plasma-surface interaction Plasma is formed + + Field electron emission Sputtering And so on ... Ion bombardment

Energy of incident ions ~8 keV (H. Timkó)

Sputtering yield (Cu on Cu) Matsunami et al. semi-empirical N. Matsunami, Y. Yamamura, Y. Itikawa, N. Itoh, Y. Kazumata, S. Miyagawa, K. Morita, R. Shimizu, and H. Tawara, in Energy Dependence of the Yields of Ion-Induced Sputtering of Monatomic Solids, IPPJ-AM-32 (Institute of Plasma Physics, Nagoya University, Japan, 1983).

Simulation E 14GV/m, 2keV Color indicates kinetic energy

Sputtering yield vs. field

14GV/m, 2keV 0.4e Color indicates charge

Sputtering or evaporation? Normalized sputtering rate (1/ps)

Sputtering or evaporation?

Modified sputtering yield

Surface features Field emitters are present on surface e.g. Protrusions may appear on an initially flat surface in the presence of a strong electric field How do these affect sputtering? A. Pohjonen et al., J. Appl. Phys. 114, 033519 (2013) E=13GV/m (void)

Insert picuture here

Surface features Field enhancement isoline h=5nm, r=1nm

Local fluence “Extreme” case with very low energy ion. Illustrates how there are many more ions hitting at the protrusion than around. Tip radius =~1nm and height = ~5nm

Local fluence Another way to look at it. Flux density is much increased around the tip (approximate radius 1nm) but goes to normal fairly rapidly with distance.

Effect of tip size With increasingly large protrusions the field distortion reaches further. r=1nm, h=5nm r=2nm, h=10nm

Effect of tip size Stronger electric field can grab faster ions Larger: h=10nm, d=4nm Smaller:h=5nm,d=2nm Stronger electric field can grab faster ions Increasing the initial energy of the ions prevent the ions from being deflected towards the protrusion. Should try to fit a function to data, but at least does not seem to be directly 1/x. Will check still.

Sputtering yield on protrusion

For the future Energy of incident ions much lower than thought See Kyrre's talk tomorrow ~10-100eV New simulations in this range First results by tomorrow? Study other geometries

Conclusions Sputtering yield in enhanced significantly when an electric field is applied Increase several hundred percent at high fields This should be considered in e.g. plasma calculations “Sputtering” yield a bit misleading: actually from field evaporation Sputtering yield is much higher from protrusions than flat surface The local flux is also increased locally around protrusions i.e. probability of hitting a protrusion (if they exist) is increased