Efeito do treinamento por ciclagem pseudoelástica nas propriedades mecânicas e funcionais de uma liga de níquel-titânio

Abstract

The use of nickel-titanium alloys in devices that undergo thermal and/or mechanical cycling while in service has been very common. For such devices to perform the desired function successfully, they must present functional and mechanical stabilities during cycling. In this context, the study of training processes is very important, since it allows knowing about the alloy cyclic behavior and when the stability of its properties is achieved, before practical use. The purpose of this study was evaluating the behavior of nickel-titanium wires, with 1mm diameter, when subjected to training by pseudoelastic cycling at a maximum strain of 6%. The influence of number of cycles, test temperature and strain rate on the alloy functional and mechanical responses was assessed. The functional properties were evaluated with regard to dissipated energy, residual strain, and forward and reverse transformation stresses. Tensile tests until rupture and three point bending tests were performed in order to analyze the mechanical properties. Furthermore, the transformation temperatures and the stability of the phases after cycling at different conditions were evaluated by differential scanning calorimetry and X-ray diffraction, respectively. The characterization of the sample as received has shown that the wire contained 51.4at.% nickel and 48.6at.% titanium and it was in the superelastic condition at room temperature (Af = -2.3°C). A tendency towards stabilization of the functional properties with increasing number of cycles was observed, whereas more significant changes were observed up to the 20th cycle. As far as the number of cycles increased, the cumulative residual strain increased, and critical stress to induce martensite as well as dissipated energy decreased, regardless test temperature and strain rate. This behavior was attributed to internal stresses caused by increased dislocation density and, possibly, by the presence of residual martensite during cycling. It was found that the effects of increasing strain rate and increasing temperature were similar, resulting in a higher cumulative residual strain and critical stress to induce martensite. This is explained by the fact that raising strain rate promotes an increase in the alloy temperature. It was also observed that the increase in number of cycles results in lower intensities of austenite diffraction peaks, which indicates a higher density of defects in the materials microstructure. However, the higher the test temperature and strain rate, the greater was the peaks intensity of the austenitic phase, suggesting that the rise in the alloy temperature is related to stress relaxation and to the stabilization of parent phase. It was not observed the presence of peaks regarding martensite at any conditions, which indicates that either the number of cycles or the maximum strain amplitude used for cycling was not sufficient to stabilize that phase. The alloy mechanical properties, evaluated after training, did not show such a significant variation in any of the cases studied.