PSI - Issue 52

Tomáš Vražina et al. / Procedia Structural Integrity 52 (2024) 43 – 51 Tomáš Vražina et. al./ Structural Integrity Procedia 00 ( 2022) 000 – 000

49

7

Regardless of the material batch, the prominent site of fatigue crack initiation was the surface of tested specimen (Fig. 6 a, b). It is expected that during the cyclic straining at room temperature irreversible plastic strain localization results in the formation of persistent slip bands (Babinský et al. 2021). Subsequently, the surface relief consisting of intrusions and extrusions is created due to the production, annihilation, and migration of point defects (Polák 2020) and serve as stress concentrators where fatigue crack emerged and advanced through the striation mechanism (I. Šulák et al. 2022). Fig. 6b shows th e fracture surface of the specimen cyclically strained with ε a = 0.5%. These relatively high strain amplitudes are characterized by multiple initiations of fatigue cracks. Phenomena such as the presence of multiple initiation sites and crack propagation through the striation mechanism are independent of the manufacturing process and occurred in both LPBF and HR materials. However, despite similarities, the fracture surfaces of LPBF material differ from those of HR material. This is mainly due to the presence of specific defects associated with the additive technology like unmelted particles (Fig. 6d), satellites (Fig. 6e), and pores (Fig. 6f). Although the contribution of damage caused by AM defects to the fatigue of LPBF samples is considered, its impact is arguable. As shown in the da/dN- ΔK diagram in Fig 6c, the slope of the LPBF curves is slightly higher than the slope of the HR curves, and the transition to the third stage, which is typical for fast crack propagation leads to the final crack, occurs earlier. This could suggest the influence of defects on the propagation of fatigue crack. These findings align with those reported by (Pei et al. 2019). It should be noted that the relative density in (Pei et al. 2019) study was approximately 99.5%, lower than the 99.86% estimated in this research. Consequently, the influence of defects, especially lack of fusion defects, which can induce the formation of microcracks and increase crack growth rate in LPBF specimens, was much more severe. The plotted dependence of crack growth rate versus the range of stress intensity factor in Fig. 6c provides only comparative information because it does not satisfy the small-scale yielding criterion as reported in (Pugno et al. 2006). According to (Hutař et al. 2017) , a relatively significant shifting of curves attributed to high strain amplitude should be observable similarly to Fig. 6c and strain amplitude ε a = 0.7%. The experimental data obtained through striation spacing assessment using the methodology proposed by (Nedbal et al. 2008) are well fitted by the Paris law, and Table 4 summarizes the constants of the Paris law for both manufacturing states of the specimen and two strain amplitudes (ε a = 0.5% and ε a = 0.7%). Nevertheless, for a more accurate description of crack development, it is necessary to take into account elastoplastic behaviour as mentioned by (Hutař et al. 2017) and (Kruml et al. 2011).

Fig. 6 a) Fracture surface of LPBF specimen with marked region of crack propagation by striation mechanism b) fracture surface of HR specimen with marked region of crack propagation by striation mechanism c) da/dN- ∆K curve of HR and LPBF material d) embedded small unmelted particle e) unmelted particle with satellite f) porosity.