Fatigue Driving Force in Superelastic Nitinol using Digital Image Correlation Measurements

Predicting fatigue life of Nitinol is critical for ensure durability of Nitinol-based cardiovascular implants. For most metals, high cycle fatigue in the elastic regime is controlled by stress. However, for superelastic materials like Nitinol, there is path dependent hysteresis during transformation regime which makes a stress measure inconvenient to predict high cycle fatigue, similar to yielding behavior of traditional metals due to plasticity. To circumvent this apparent difficulty associated with stress-based fatigue prediction, strain-based techniques have been employed until now to estimate fatigue life of Nitinol to include the effect of partial transformation. In previous research, we have demonstrated the ambiguity and lack of consistency between continuum mechanics based finite element simulation and actual strain measurement in the mixed phase regime using digital image correlation (DIC). These discrepancies between FEA and DIC strains in the transformation region has been attributed to the “law of averages” approach of modeling transformation used in most finite element codes. This divergence has been consistently observed in specimens for both tensile and bending loading which reiterates the deficiency of the sole adoption of FEA strains to drive stain-based fatigue life assessment especially understanding the effects of mean strains. In this study we have further investigated this deviation for a flexure Nitinol coupon (C-specimen) and re-examined the DIC measured strains for previous published data. These DIC strain measurements can be used not only to compare and validate the FEA simulation, but also to explore the true driving force for fatigue initiation in the mixed phase regime.