PSI - Issue 28

Costanzo Bellini et al. / Procedia Structural Integrity 28 (2020) 2283–2290 Author name / Structural Integrity Procedia 00 (2019) 000–000

2285

3

Other works of Furgiuele and Maletta (2010) and Sgambitterra et al. (2016) showed the importance of temperature on the behaviour of SMAs, proponing approaches that could be able to correctly describe the effect of austenite and martensite on the crack propagation using Clausius–Clapeyron relation. However, these approaches must be confirmed by experimental results. Di Cocco et al. (2014) showed that fatigue cycles affect not only the mechanical response of the material but also the transformation of the phases. These variations are much more pronounced during the first fatigue cycles and postulate a different variation of the microstructure with the same deformation. As the number of cycles increases, these variations tend to decrease significantly, until they are imperceptible between a cycle and the next one. This behaviour was observed starting from about the tenth cycle, with the used measurement techniques. The techniques used to evaluate the quantity of austenite and martensite proposed in the literature are different, starting from common X-ray diffractometric analysis to calorimetric measurement approaches. Mellor et al. (2012) proposed an innovative approach, based on micro indentation test results and a calculation model adopted to quantify the content of phases. Moreover, the proposed model has been used to analyze the effects of temperature on the indentation response of the alloy. In the present work, using the diffraction results obtained in two different works of Carpinteri et al. (2018) and Di Cocco et al. (2014) and the model proposed by Mellor et al. (2012), the microstructure evolution of a NiTi equiatomic alloy characterized by a pseudoelastic behaviour has been evaluated, considering also the dimension of the elementary cell. The result could explain the presence of the plateau in multiple SMAs or the marked slope variation of the stress strain curve during the austenite-martensite transformation. 2. Materials and methods The material used for this investigation has been obtained by under-vacuum fusion of nickel and titanium metals (pure metal used in laboratory) in order to obtain an equiatomic alloy characterized by a pseudo-elastic behaviour. The melting furnace was equipped with a yttrium oxide crucible in order to limit the release of other elements in the melting process. The casting was done by centrifugal force in a graphite mould. In fig. 2 some details of the furnace and its equipment are shown.

Fig. 2. Details of furnace, crucible and mould.

The ingots were obtained by casting the melted metal in the furnace, then mini dog bone specimens were machined from the ingots (Fig. 3) and clamped on a patented tool able to impose constant deformations. This tool can be mounted in an x-Ray chamber in order to perform the diffraction test on the specimen in a deformed condition. The spectra analyses have been performed using an iterative algorithm considering at least three peaks to interpolate. Finally, the quantity of phases has been calculated using the mean peak intensity model.