Abstract EN:

Metal Additive Manufacturing also referred to as 3D printing has flourished rapidly into sectors such as aerospace and automotive, where high strength-to-weight ratio and defect-free parts are key requirements. Until today only a few aluminum alloys like AlSi10Mg and AlSi7Mg are manufactured by laser powder bed fusion (L-PBF) [1]. Unfortunately, structural alloys from the 6XXX-series (Al-Si-Mg) and 7XXX-series (Al-Zn) are frequently reported to be crack-sensitive under high cooling rate conditions typical of the L-PBF process [2,3]. In the literature, few technical solutions have been suggested to overcome hot cracking issues. Martin et al. [4] proposed to add nano-particles to promote grain nucleation and thus refine grain size. Others suggest modifying the alloy composition by adding elements like Si [5] or rare-earth elements like Sc in Scalmalloy®. However, there are still debates regarding the mechanism leading to hot cracks in parts made of 6061-grade built by L-PBF. This lack of in-depth understanding of the root causes of hot cracking is an impediment for designing engineering parts for safety-critical applications.The mechanism at the origin of cracks has been identified as solidification cracking based on the observation of the fracture surface of as-built parts where a dendritic morphology gives evidence of the presence of liquid films trapped in the interdendritic space. Also besides, our experimental outcomes based on EBSD characterization, demonstrates that cracking occurs only along columnar grain boundaries of higher misorientation (>20°). We rationalize this cracking along high misorientation grain boundaries behavior using the Rappaz model based on the critical coalescence undercooling [6]. Thus, the critical coalescence undercooling can be estimated as a function of misorientation.The Rappaz-Drezet-Gremaud (RDG) hot cracking criterion [7] is then applied using inputs that comply with L-PBF processing conditions. First, the solidification path of the 6061-alloy is calculated relying on the Scheil-Gulliver assumption. Second, the critical coalescence undercooling is included in the RDG model by modifying the lower integration limit to account for the fact that liquid films are stables at lower temperatures along high misorientation grain boundaries. Finally, the thermal gradients (G) and solidification velocity (v) required into the RDG are estimated with the help of thermal simulations using the Rosenthal analytical model to get values typical of the L-PBF process.Our findings for 6061 alloy show that the existence of stable liquid films is linked to grain boundary misorientation, which causes a sudden increase in pressure drop leading to cracking. We also evaluate thanks to our modeling approach:• the effect of the processing conditions, namely the first-order melting parameters (laser power and scanning speed) on the thermal gradient and solidification velocity inferred from thermal simulations, on the hot tearing sensitivity. This led us to an understanding of the required (G, v) and therefore the required laser power and speed to decrease the cracking susceptibility and propose improvements to process the 6061 alloy using L-PBF.• the effect of solute content modification on the cracking sensitivity. This can be further used as guidelines to suggest chemical composition modification of the 6061 Al-alloy to improve its processability by L-PBF.