AVS 49th International Symposium
    Plasma Science Wednesday Sessions
       Session PS+TF-WeP

Paper PS+TF-WeP6
Eliminating the Hysteresis Effect for Reactive Sputtering Processes

Wednesday, November 6, 2002, 11:00 am, Room Exhibit Hall B2

Session: Plasma Etching & Deposition
Presenter: S. Berg, Uppsala University, Sweden
Authors: T. Nyberg, Uppsala University, Sweden
S. Berg, Uppsala University, Sweden
U. Helmersson, Linkoping University, Sweden
Correspondent: Click to Email
Despite that reactive sputtering has existed for more than 50 years, increasing the pumping speed to unrealistically high values is so far the only reported way of eliminating the hysteresis effect for planar magnetrons. The cause of the hysteresis effect is a complex interaction between target sputter erosion and the gettering of the reactive gas on the target and coated surfaces. These relations are not yet fully understood. By computer process modeling, however, we have developed a theoretical model capable of predicting the complex correlations between these involved parameters. By process modeling it is possible to carry out ”virtual processing” to predict processing results for unexplored ways of carrying out processes. By reducing the size of the target sputter erosion zone below a critical value, simulations predicted that it should be possible to completely eliminate the hysteresis. This was also experimentally verified for the reactive sputtering of Al in an argon/oxygen atmosphere. Moreover, there is almost no loss in compound deposition rate as compared to sputtering from a traditionally designed target. Notice that an increase of the total target current using a large sputtering erosion zone will not eliminate the hysteresis. The fundamental explanation to this behaviour as well as experimental verification will be presented.