Wear monitoring of protective nitride coatings using image processing

A double-layer model system, consisting of a thin layer of tribological titanium aluminum nitride (TiAlN) on top of titanium nitride (TiN), was deposited on polished 100Cr6 steel substrates. The TiAlN top-coatings were exposed to abrasive wear by a reciprocating wear process in a linear tribo-meter with up to 105 repetitive cycles, eventually leaving the embedded TiN signal layer uncovered at the bottom of the wear scar. The worn surface was characterized by subsequent image processing. A color detection of the wear scar with the exposed TiN layer by a simple optical imaging system showed a significant increase up to a factor of 2 of the relative color values from the TiAlN top layers to the embedded TiN signal layers. This behavior agrees well with the results of optical reflectance detection experiments with a red laser on the same system. Thus we have demonstrated that image processing by color detection is a potential technique for early warning or determination of residual thickness of tribological tool coatings prior to complete wear.