PSI - Issue 66

14

A.R. Pelton et al. / Procedia Structural Integrity 66 (2024) 265–281 Pelton/ Structural Integrity Procedia 00 (2025) 000–000

278

Crystallographic characterization of the sheet suggested elongated grains around 20 µm in size with preferential alignment of the crystallographic orientation [001] in the rolling direction (Brambilla, Berti et al. 2024). Complementary tension-tension testing was also conducted on “wires” that were extracted from superelastic Nitinol tubing with similar texture. Analysis of the defect sizes that correspond to 50%, 95%, and 5% percentile values of the defect size distribution are equal to 1.205 μ m, 2.818 μ m, and 0.299 μ m as shown in Figure 12. Clearly, these defect sizes represent small cracks rather than large cracks that were reported previously (Robertson and Ritchie 2008). A recently published paper attempts to describe the mechanisms with ultrahigh cycle fatigue ( ≥ 10 8 cycles) in superelastic Nitinol wire by rotary bend fatigue (Roiko, Cook et al. 2025). This paper is an extension of their first publication where they reported on rotary bend fatigue results of superelastic Nitinol wires out to a billion cycles (Weaver, Sena et al. 2023). In the more recent publication, the investigators paid special attention to the fracture surfaces, in particular to the region adjacent to the inclusion initiation site. They introduced a term based on the observation of a different morphology in fractures that occurred in specimens ≥ 10 7 cycles, namely, the “Region of Reduced Roughness” (RRR). The authors compared the RRR to the growth and size a fractographical feature (Optical Dark Area) in steels (based on extensive work by Murakami and co-authors (Murakami and Beretta 1999, Nagata and Murakami 2003, Murakami, Kanezaki et al. 2008, Roiko and Murakami 2012)). Figure 13 shows an example of such a superelastic Nitinol fracture surface with the RRR near the NMI with profilometry height data (left) and corresponding SEM image (middle) of a specimen that was subjected to ε a = 0.40 % and fractured after 1,573,210,491 cycles. Despite contrary examples in the literature for ultrahigh cycle fatigue of steels (Bathias 2014), fatigue initiation in those superelastic Nitinol samples were not necessarily at subsurface inclusions but were at most within 2µm of the surface (Roiko, Cook et al. 2025). In accord with previous literature (for example, (Roiko and Solin 2014)), the authors calculated the stress intensity factor with the following equation: Δ K defect = 0.65• Δσ • √ (  √ area defect ), where Δσ is the applied stress range, and √ area defect is the square root of the cross-sectional area of the defect or inclusion normal to the applied stress range. Figure 13 (right) graphs the cumulative probability distribution of the ultra-high cycle fatigue RRR and the corresponding inclusion sizes as measured by SEM.

Fig. 13: Profilometry height data (left) and corresponding SEM image (middle) of a specimen that was subjected to ε a = 0.40 % and fractured after 1,573,210,491 cycles. These images and accompanying analyses define the “region of reduced roughness”, RRR, for the samples that fracture at cycles ≥ 10 7 . The usefulness of this RRR values is shown in Fig. 13 (right) with both the inclusion size and RRR on the abscissa and probability on the ordinate. After (Roiko, Cook et al. 2025).