PSI - Issue 33

V. Di Cocco et al. / Procedia Structural Integrity 33 (2021) 1035–1041 Author name / Structural Integrity Procedia 00 (2019) 000–000

1036

2

1. Introduction Shape Memory Alloys (SMAs) are a class of alloys characterized the ability to recover the original shape. It is due to a phase transition process. This phase transition, that is defined as Thermoelastic Martensitic Transformation (TMT), happens between the parent phase, that is called austenite, and the product phase, that is called martensite, and it is due to two different effects: the Pseudoelastic Effect (PE), connected to stress state, and the Shape Memory Effect (SME), associated to temperature changes (Natali et al.). The transformations happen without recrystallization and the direct one, that is from austenite to martensite, is caused by mechanical deformation. As concern the transformation from martensite to austenite, that is the inverse one and it allows the recovery of the initial shape, it can be induced by heating if the part temperature is higher than the critical value (SME), otherwise it is achieved by removing the applied load (PE) (Brotzu et al.). In the latter case there is not a linear connection between the strain and the stress, because three diverse steps can be observed in the stress-strain curve: the first one is characterized by a linear elastic load increase and the material is constituted by the austenitic phase; in the second one there is a deformation increase without load increase coupled to the phase transformation; finally, in the third step the material is constituted by martensite and there is another linear elastic load increase (Di Cocco et al.). Near equiatomic nickel-titanium alloy (NiTi) is one of the most common SMA is the, that is employed for several applications, as joining valves, actuators, smart sensors, and commercial gadgets. Moreover, due to the high biocompatibility, this alloy is largely used in medical applications to produce orthopaedic prosthesis, filters, and endoscopic devices. The pseudolasticity makes this alloy suitable for the manufacturing of self-expanding stents and wires for orthodontics (Kuribayashi et al.). Due to their versatility, several research works were carried out for determining the thermomechanical characteristics of NiTi and different numerical models were proposed to simulate their behaviour. Nevertheless, classical elastic or elasto-plastic models cannot be used for this purpose due to the phase transformations, so specific models are needed to correctly consider both stress induced and temperature induced transitions. The crack propagation behaviour of NiTi is different if compared to other metals because it is influenced by the phase transformation (Shimamoto et al.). The crack propagation in NiTi alloys is a topic that has aroused interest among scientists, as witnessed by several studies that analysed this phenomenon for both static (Gollerthan et al., Furgiuele et al.) and cyclic loading (Gall et al., Robertson et al.). Moreover, some numerical models, coupling plasticity theory with FEM (Wang et al.), and analytical models, constructed on the linear elastic fracture mechanics (Baxevanis et al., Maletta et al.), were introduced. Recently, new investigations have been carried out by several researchers. Iasnii and Yasniy studied the influence of the loading ratio on the low cycle fatigue behaviour of a NiTi alloy by employing different criteria based on stress, strain, and energy. Dornelas et al. investigated the fatigue of NiTi alloy through experimental tests, considering both phase transformation and plasticity; then, the experimental data were used for generating a predictive model, and the fatigue behaviour was determined by considering the different characteristics of the two phases. Hashemi et al. introduced a rate dependent model to delineate the fatigue curves of NiTi springs by considering the effect of mean applied displacement and strain rate. The model was validated by comparison with experimental tests and the load frequency was found to affect the fatigue behaviour. Pereira et al. carried out experimental fatigue test to correlate the maximum number of cycles and the strain with the strength decrease; moreover, the introduced analytical model was validated through numerical simulations. Frost et al. proposed a methodology to study the martensite formation during the cyclic loading of a NiTi spring by combining experimental tests and numerical modelling. Furthermore, they found an effect of the stress asymmetry on the phase distribution. Vantadori et al. studied the fatigue behaviour of a NiTi alloy by tensile tests coupled to X-ray diffraction analysis to determine the phase content; in such a manner, an analytical model suitable to delineate the alloy behaviour was introduced. Bellini et al. proposed a numerical model to correlate the phase transition with the mechanical behaviour of the material, and it was validated through numerical tests. Allegretti et al. applied different models based on material strain to predict the fatigue life of NiTi stents: the Von Mises model overestimated the failure risk, while the critical plane one were more precise and reliable. Bagheri et al. carried out experimental tests on NiTi specimens produced through additive manufacturing process and compared their mechanical properties with those of wrought specimens. A shorter fatigue life was found for the former specimens, caused by the presence of process induced voids. A fatigue limit decrease was found for a stress level quite far from the phase transition onset; moreover, the temperature influenced the fatigue behaviour too. Di Cocco and