All Title Author
Keywords Abstract


Multi objective optimization of wear resistant TiAlN and TiN coatings deposite by PVD techniques

Keywords: Computational material science , Thermal stresses , FEM , Multi-objective optimization

Full-Text   Cite this paper   Add to My Lib

Abstract:

Purpose: The goal of this paper is to determine, the optimal layer thickness of deposited coatings, in respect of thermal strain and stresses.Design/methodology/approach: For physical modelling purposes Cr, TiN and TiAlN layers were treated as a continuous medium, so the physical phenomena, occurring in the coating, are modelled based on a classical theory of stiffness. Computer model of the object (coating + substrate) describing strains and thermal stresses states in layers, after deposition process, was created using FEM method.Findings: The decisional objectives, based on various stresses fields in deposited coating, were defined. The set of optimal TiAlN and TiN layer thickness, in respect to created decision objectives was determined. Also method of optimal solutions set analysis, based on multidimensional, Euclidean metric was created.Research limitations/implications: There is a need to consider creation of a certain class of selection functions, as a standard, which will help to choose the optimal set of solutions - obtained in different multi objective optimization procedures. Of course, new and more detailed physical and mathematical models of the PVD processes are required.Practical implications: Proposed multi objective optimization procedure will become a component of the PC software in future, which will make design process of hard, wear resistant coatings architecture possible.Originality/value: Insertion of the base layer, below TiAlN and TiN tiers, was proposed, whose occurrence is reflected by the continuous change of the physical and chemical properties, across the coating thickness. Also method of optimal solutions set analysis, based on multidimensional, Euclidean metric was created.

Full-Text

comments powered by Disqus