Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Three dimensional plasma simulations with a parallelized particle-in-cell Monte Carlo approach

: Siemers, M.; Pflug, A.; Szyszka, B.

Society of Vacuum Coaters -SVC-, Albuquerque/NM:
Society of Vacuum Coaters. 49th Annual Technical Conference 2006. Proceedings : April 22 - 27, 2006, Washington, D.C. USA
Albuquerque: SVC, 2006
Society of Vacuum Coaters (Annual Technical Conference) <49, 2006, Washington/DC>
Fraunhofer IST ()

Improving the theoretical understanding of the reactive magnetron sputtering process regarding facility layout and process control is essential for gaining access to high-precision film properties in consideration of homogeneity and reproducibility. Especially insights about plasma physics and target interaction within a magnetron discharge are required to derive reliable conclusions about dynamic process behavior and underlying parameter correlations on a macroscopic scale.
In this work a parallelized three dimensional plasma simulation environment is introduced using the classical hybrid approach of combined Particle-in-Cell and Monte-Carlo simulation (PIC-MC) to handle charged particle movement as well as molecular interactions. Electric and magnetic fields are computed seperately using the Finite Difference Method (FDM) and the Boundary Element Method (BEM) respectively. For an increased flexibility in geometry decomposition and for the simulation of more complex boundary particle interactions, triangle meshed freeform objects can be mapped both into the field compulation and the molecular interaction domain.
An exemplary PIC-MC simulation on a planar magnetron cathode is demonstrated taking into account three dimensional field geometry and spatial molecular dynamics. In contrast to simplified one- or two-dimensional models this holistic approach reveals a more detailed perception about plasma parameters and their coupling behavior along the target racetrack. This includes visualization of geometric target anomalities such as inhomogeneous target erosion and poisoning. The results obtained from the PICMC simulation can be further used for refinement of a heuristic plasma impedance model.