No abstract

Depth profiling has been performed by using Auger electron spectrometry and X-ray photoelectron spectrometry in combination with Ar-ion sputtering. The data obtained by both surface-analytical methods have been evaluated by means of factor analysis and partly by applying artificial neural network in order to determine the compositional layering of different thin-films such as TiNx on Ti, Cr2O3/CrN sandwich layer, and copper oxide on Cu. Both multivariate statistical methods applied to the same data sets lead to results that agree well within statistical deviations provided that the structure of the artificial neural network is constructed appropriate to the actual problem.

Abstract. Chromium nitride and chromium oxide were deposited by arc PVD and APS on steel substrates to be worn in a testing set up simulating the abrasion process of a screw rotor compressor. The damage of CrxN is more severe than that of Cr20 3. The abrasive particles were collected on glassy carbon with a Batelle impactor and analyzed by AES and REM. Particles from the Cr~N coating are totally oxidized, in contrast to Particles from Cr20 3 which are not chemically influenced. No oxidation can be detected on the surface of the treated CrxN coating. The Cr20 3 coating is not affected either.