Multilayer nitride coating performance optimized by an artificial neural network approach

Abstract

One of the most important problems occurred in many industries is due to friction and wear process. Over the years, minimizing friction and controlling wear is one of the difficult tasks for the researchers. Both properties can be minimized by the application of adequate coating technology. Many coating deposition technologies have been employed to limit friction and wear but only few succeeded, those are directly affected by the nature of the material under investigation and process parameters. A suitable coating strategy varying from single layer to multilayer should be applied to the materials whose superficial properties such as low friction, improved wear resistance, and adhesion are the prime interest. Multilayer coatings possess high hardness, ductility and fracture strength compared to single layer coatings. The advantageous properties of these multilayers can be preciously tailored according to specific application. For this purpose Physical Vapour Deposition (PVD) coatings have been developed considerably due to increasing industrial demands. In the present research, friction and wear study of multilayer PVD-nitride coating deposited on tool steel by unbalanced reactive magnetron sputtering technique have been discussed. Later on an Artificial Neural Network approach was used to predict the tribological properties of multilayer nitride films. Bias voltage, total gas flow rate, lap, time, velocity and load were considered as controllable factors. The regression and performance curve analysis is used to assess the optimized outcome of deposited film properties such as friction and wear. The analyzed results shows that experimented and predicted values are in a good agreement.