Adhesion scratch testing - A round-robin experiment

Abstract

Six sets of samples, TiN coated by chemical or physical vapor deposition methods (CVD or PVD) onto cemented carbide or high-speed steel (HSS), and TiC coated by CVD onto cemented carbide have been scratch tested using three types of commercially available scratch adhesion tester. With exception of one cemented carbide set, the reproducibility of the critical loads for any given set with a given stylus is excellent, about + or - 5 percent, and is about + or - 20 percent for different styli. Any differences in critical loads recorded for any given sample set can be attributed to the condition of the stylus (clean, new, etc.), the instrument used, the stylus itself (friction coefficient, etc.), and the sample set itself. One CVD set showed remarkably large differences in critical loads for different styli, which is thought to be related to a mechanical interaction between stylus and coating which is enhanced by a plastic deformability in the film related to the coating microstructure. The critical load for TiN on HSS increases with coating thickness, and differences in frictional conditions led to a systematic variation in the critical loads depending on the stylus used.