PSI - Issue 83

Ahmad Issmail et al. / Procedia Structural Integrity 83 (2026) 229–238

230

1. Introduction Owing to its high resolutions, laser-based powder bed fusion (PBF-LB) has emerged as a key additive manufacturing (AM) process for producing lightweight topology-optimized components for a wide range of industrial applications. Among the available alloys, AlSi10Mg is being increasingly employed in aerospace and automotive applications due to its good processability and high specific strength [1]. However, the layer-by-layer process combined with the complex geometry produces parts with unique thermal history-dependent properties, including micro- and mesostructural anisotropy, process-induced defects, and irregular surface textures that significantly affect mechanical and fatigue properties [2, 3]. In the as-built condition, the high thermal gradients and high cooling rates result in a unique microstructure that consists of a supersaturated solution in α -Al cells surrounded by eutectic intercellular Si-network formed within the melt pools (MPs) that are imprinted in the mesostructure due to rapid solidification [4]. The direction of the heat dissipation with respect to the build orientation dictates the solidification patterns, introducing orientation-dependent microstructure, thereby anisotropic mechanical and fatigue behavior [5, 6]. Build orientation also dictates surface roughness, which has the primary negative impact on the fatigue behavior in as-built surface conditions [7 ,8]. In addition, the as-built specimen size affects thermal history, resulting in alterations in the heat dissipation magnitude and direction, heat accumulations, and MP morphology, leading to impacts on microstructure, surface texture, and consequently mechanical and fatigue properties [9–12]. To eliminate micro and mesostructural sources of anisotropy, post-process homogenizing heat treatments are performed and result in the decomposition of the thermal history imprints through Si diffusion from the α -Al cells and its spherization through aggregation with the Si of the disrupted Si-network [13]. This mechanism is associated with the improvement of ductility and fatigue behavior at the expense of strength [7, 14, 15]. The fact that fatigue cracks most often initiate from the rough as-built surface of AM specimens, and that cracks subsequently propagate along MP boundaries (MPBs) [8], means that the roughness profile and the MP orientation synergically determine the fatigue life variation of as-built AM parts with size and build orientation. Therefore, a better understanding of the crack initiation-induced variation of fatigue life with specimen geometry requires that the growth induced anisotropy must be taken out of the equation. For this reason, this work evaluates the tensile and fatigue behavior of four heat treated specimen categories with different combinations of specimen size and build orientation, previously analyzed under a stress-relieved condition [16].

Nomenclature S|V

Standard-sized specimen with vertical build orientation Standard-sized specimen with 45° build orientation Standard-sized specimen with horizontal build orientation Miniature-sized specimen with 45° build orientation Miniature-sized specimen with horizontal build orientation

S|45°

S|H

m|45° m|H UTS EaB S max S F,lim MP/s MPB/s N f

Ultimate Tensile Strength Elongation at Break [%] Maximum stress Fatigue limit at 2×10 Number of cycles to failure 6 cycles

Melt pool/s

Melt pool boundary/s

HT SR

Heat treatment Stress relief/ed