PSI - Issue 66

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

272 8

analyses are useful as a complement to a life prediction approaches in order to understand mechanisms related to crack initiation and crack path in medical devices with strut dimensions of a few hundred micrometers. Damage tolerant analysis can provide life prediction and safety of a device based on the presence of flaws and/or defects. As such, determination of parameters such as the (small or large crack) fatigue crack growth threshold , Δ K th , or and determination of critical flaw size based on the service load could be used for device inspection criteria. McKelvey and Ritchie conducted damage tolerance studies on disk-shaped compact tension DC(T) samples that were electro-discharge machined with dimensions of 31 mm in width and 9 mm in thickness from a 41.3mm diameter round bar medical-grade superelastic Nitinol (McKelvey and Ritchie 1999). Samples were cycled at 37  C at a load ratio ( R ) equal to 0.1, where R is defined as the ratio of the minimum to the maximum stress intensities in the loading cycle ( i.e ., R = K min / K max ). Figure 6 is a plot of the fatigue-crack growth rates for Nitinol compared with other biomedical metallic alloys, namely, stainless steel, pure titanium, Ti-6Al-4V, and a CoCr alloy. This comparative data indicates that the fatigue threshold ( ∆ K th ~ 2 MPa √ m) was the lowest and the crack-growth rates were the fastest in Nitinol. After (McKelvey and Ritchie 1999). A follow-up study to the synchrotron x-ray µdiffraction investigation discussed above (Barney, Xu et al. 2011) was conducted in order to understand the effects of microstructure on a growing crack in superelastic Nitinol (Robertson, Mehta et al. 2007). Ultrahigh spatial resolution (~1µm 2 ) synchrotron X-ray microdiffraction was combined with fracture mechanics techniques to measure directly in situ three-dimensional strains, phases and crystallographic alignment ahead of a growing fatigue crack (100 cycles in situ ) (Robertson, Mehta et al. 2007). C(T) specimens were laser machined from flattened Nitinol tubing and then pre-cracked ex situ at ∆ K=3 MPa √ m. The pre-cracked specimens were then loaded on the beamline with a custom-designed in situ micro-tensile loading device to the desired cyclic mode I stress intensity of 7.5–15 MPa √ m. These authors demonstrated that the strain distribution and transformation did not follow the expected peanut shaped front that is predicted from linear–elastic fracture mechanics models. Instead, grains with, for example, <100> orientations could suppress the transformation even when surrounding grains had already undergone the transformation. These observations of non-uniform distributions in strain at the tip of a growing crack, as shown in Figure 7, are similar to the effects of microstructure

Fig. 6: Comparison of fatigue-crack growth rates for various biomedical metallic alloys, including Nitinol, stainless steel, pure titanium, Ti-6Al 4V, and a CoCr alloy. This comparative data indicates that the fatigue threshold ( ∆ K th ~ 2 MPa √ m) was the lowest and the crack-growth rates were the fastest in Nitinol. After (McKelvey and Ritchie 1999).