Study of volumetric energy density limitations on the IN718 mesostructure and microstructure in laser powder bed fusion process

Abstract

The optimization of process exposure parameters is needed to guarantee the desired meso- and microstructural properties of parts manufactured by laser powder bed fusion (LPBF). The volumetric energy density combined parameter (VED) has been used as a porosity and track’s continuity optimization parameter; however, some studies show high variability on defects at the same VED values regardless of the parameters combination to achieve it. This work compares an optimal region in terms of the VED value with a multiparametric optimal region in terms of laser power, scan speed, and hatch distance for Inconel 718 cylinders fabricated by LPBF. Both regions were described in terms of the mesostructural defects minimization and microstructure. A factorial design was carried out to study the individual effect and the interactions of the VED parameters (laser power, scan speed, hatch distance) on the molten pool depth, porosity, mesostructure defects, and microstructure. Porosity was measured by X-ray tomography, and the meso- and microstructure were analyzed based on optical and scanning electron microscopy. The porosity level determined three working regions over the studied experimental window: porosity ≥1%, porosity ≤1%, and overheating. Under the studied experimental conditions, the hatch distance shows the most significant impact on porosity, followed by laser power. The confidence range of VED for the porosity optimization parameter was between 55.8–147.2 J/mm3, which is smaller than the region of porosity ≤1% described in terms of the individual factors. Since at VED < 55.8 J/mm3 some experimental points are in the operable optimal window. At lower VED (47.3 J/mm3) a multiparametric optimization increased the LPBF operable optimal window. The results of binding-fault porosity demonstrated that the weight of every exposure parameter does not match the standard VED mathematical definition. These results reveal that the VED parameter could not be used as a global optimization parameter; however, it can be used as the first approach for minimization of defects.