Panwisawas_et_al(2017)ComMaterSci-Accepted.pdf (1.64 MB)
Download file

Mesoscale modelling of selective laser melting: Thermal fluid dynamics and microstructural evolution

Download (1.64 MB)
journal contribution
posted on 07.04.2020, 12:33 by C Panwisawas, C Qiu, MJ Anderson, Y Sovani, RP Turner, MM Attallah, JW Brooks, HC Basoalto

In this paper, an integrated computational materials science approach for selective laser melting (SLM) at the mesoscale is presented. A particle dropping model was developed to simulate the representative powder-bed particle distribution of a measured titanium alloy powder. Thermal fluid flow and resulting microstructural evolution of a set of laser scanned single tracks with different powder layer thicknesses and scanning speeds during SLM were also studied using both computational and experimental approaches. The simulated powder particle distribution was found to be consistent with experimental measurement. The thermal fluid flow model predicts that single laser scanned tracks become increasingly irregular-shaped with increased powder layer thickness and increased laser scanning speed. These findings were reinforced by scanning electron microscopy analysis. The more dispersed dissipation of the localised heat for thicker powder layers is understood to cause increased melting and evaporation. This can lead to increased Marangoni force and recoil pressure which in turn destabilises the melt flow. The use of an argon atmosphere speeds up the solidification process when compared with air but does not affect the morphology of single tracks significantly. The predicted microstructure was consistent with the electron backscattered diffraction data. The microstructure-based modelling methodology considering the representative powder size distribution provides a good predictive capability for the laser-powder interaction behaviour, surface structure and porosity development.

History

Citation

Computational Materials Science Volume 126, January 2017, Pages 479-490

Version

AM (Accepted Manuscript)

Published in

Computational Materials Science

Volume

126

Pagination

479 - 490

Publisher

Elsevier BV

issn

0927-0256

Acceptance date

09/10/2016

Copyright date

2016

Available date

27/10/2016

Publisher version

https://www.sciencedirect.com/science/article/abs/pii/S0927025616305079

Language

en

Usage metrics

Categories

Exports