Effect of processing parameters on titanium nitrided surface layers produced by laser gas nitriding

Experimental investigations based on Response Surface Method (RSM) were carried out to study the influence of laser power, nitrogen flow rate, and scanning speed on the microstructure, depth, hardness, and cracking of a set of titanium nitrided surface layers produced by laser gas nitriding. Laser powers ranged between 1 and 5 kW, scanning speeds 5 to 20 mm/min, and nitrogen gas flow rate ranged from 500 to 4000 l/min. The aim was to use RSM in the design of experiment to find suitable processing parameters which produce deep and crack free nitrided layers with a high surface hardness. Optical microscope, scanning electron microscope equipped with energy dispersive spectroscopy (EDS) analysis, and X-ray diffraction were used to characterize the microstructure and composition of the nitrided layers. Microhardness at a distance of 0.15 mm from the surface for all tracks was measured. The results showed that laser melting of titanium surface in a nitrogen containing atmosphere has led to the formation of a nitrided layer characterized with a strong convective flow and of a dense structure of TiN dendrites heterogeneously distributed. The TiN dendrites, which formed either directly from the melt or as a result of the peritectic reaction L + alpha Ti -> TiN were of various sizes and shapes and distributed non uniformly. The volume fraction of TiN dendrites in the melted zone is a function of processing speed and power being higher at slower speed and high power. The convectional flow not only affects the surface quality but also leads to effective nitrogen transport to a deeper region. The formation of TiN significantly increases the microhardness of the surface but it makes the surface rough. The optimum process parameter settings which were determined statistically in terms of power, scanning speed and nitrogen gas flow rate were found to be 2.8 kW laser power, 5 mm/s scanning speed and 2000 l/h nitrogen flow rate which would result in a maximum microhardness of approximately 1900 HV0.15. (C) 2012 Elsevier B.V. All rights reserved.