Particle-based simulation of atom and ion transport in HiPIMS: effect of the plasma potential distribution on the ionized flux fraction

We present a three-dimensional particle-based computer simulation of high-power impulse magnetron sputtering (HiPIMS) discharges which enables us to simulate the transport of atoms and ions in the discharge and the corresponding plasma parameters. The simulation requires a definition of the plasma p...

Full description

Saved in:
Bibliographic Details
Published in:Plasma sources science & technology 2023-03, Vol.32 (3), p.35007
Main Author: Kozák, Tomáš
Format: Article
Language:English
Subjects:
DSMC
HiPIMS
ion return probability
ionized fraction
magnetron sputtering
plasma simulation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a three-dimensional particle-based computer simulation of high-power impulse magnetron sputtering (HiPIMS) discharges which enables us to simulate the transport of atoms and ions in the discharge and the corresponding plasma parameters. The simulation requires a definition of the plasma potential and electron density distribution (not calculated self-consistently), for which parametric analytical formulae were devised. A numerical algorithm is used to constrain the simulation by an experimental target current waveform, which ensures that the simulation results are closely tied to the experimental discharge conditions. Simulations of a HiPIMS discharge with Ti target show the capability to calculate the spatial distributions of target material atoms and ions and also to quantify the process-gas rarefaction. We evaluated, among others, the ion return probability and the ionized fraction of the target material flux onto the substrate for various values of the potential difference across the magnetic presheath in front of the target racetrack, which is responsible for attracting most of the plasma ions towards the target. It is shown that this parameter of the plasma potential distribution strongly affects the ion return probability and, thus, it must be known quite precisely to reliably predict the ionized flux fraction on the substrate. Other parameters, such as the composition of the ion flux onto the target are less sensitive. The simulation can be run in a reasonably short time and can easily be extended by adding more plasma species (excited states or doubly ionized species) and their interactions.
ISSN:0963-0252
1361-6595
DOI:10.1088/1361-6595/acc549