Paper SE2-MoM6
Effects of HiPIMS Plasma Transport on Thin Film Deposition
Monday, November 9, 2009, 10:00 am, Room C4
Session: |
Pulsed Plasmas in Surface Engineering |
Presenter: |
D. Lundin, Linköping University, Sweden |
Authors: |
D. Lundin, Linköping University, Sweden N. Brenning, Royal Institute of Technology, Sweden M.A. Raadu, Royal Institute of Technology, Sweden U. Helmersson, Linköping University, Sweden |
Correspondent: |
Click to Email |
A new exciting development of magnetron sputtering was achieved when introducing high power impulse magnetron sputtering (HiPIMS). HiPIMS is one of the most promising improvements of common IPVD techniques and is already making its way to industrial applications. The HiPIMS plasma generates large quantities of ions of the sputtered material due to a high plasma density, but also acceleration of the ions increasing the bombardment of the growing film without using a substrate bias voltage. Also observed is a lower deposition rate for HiPIMS than that obtained for conventional DC sputtering, using the same average power. In order to optimize the process, controlling ion acceleration and increasing deposition rate, the mechanisms for transport of charged particles in this type of plasma need to be known. In the present work, it is shown that the electron mobility across the magnetic field is enhanced by typically an order of magnitude during the HiPIMS discharge compared to DC magnetron sputtering. This cannot be explained by classical theory of diffusion and electrical conductivity or Bohm diffusion. The transport is directly reflected by an anomalously low azimuthal-to-discharge current ratio, Jφ / JD = 2. On the microscopic scale, the anomalous transport can be shown to be mediated by observed azimuthal electric field oscillations in the lower hybrid range. Furthermore, new insights from experimental data and plasma discharge modeling will be presented, which show that a large fraction of the ionized species are attracted back towards the target, either by electric fields in the bulk of the plasma, or by the stronger local fields in the cathode sheath. In this context, it is demonstrated that the effect of the anomalously high electron mobility to reduce the bulk E field is important to understand and control. The study also verifies that the resistive friction force, Fi,φ, associated with the anomalous resistivity, can accelerate the ions azimuthally, as is shown both indirectly from changes in the deposition patterns, and directly by mass spectrometry.