Salta al contenuto principale
Passa alla visualizzazione normale.

GIUSEPPE PITARRESI

Investigation of the thermomechanical response of cyclically loaded niti alloys by means of temperature frequency domain analyses

Abstract

Nickel–Titanium (NiTi) shape memory alloys subjected to cyclic loading exhibit reversible temperature changes whose modulation is correlated with the applied load. This reveals the pres-ence of reversible thermomechanical heat sources activated by the applied stresses. One such source is the elastocaloric effect, accounting for the latent heat of Austenite–Martensite phase transfor-mation. It is, however, observed that when the amplitude of cyclic loads is not sufficient to activate or further propagate this phase transformation, the material still exhibits a strong cyclic temperature modulation. The present work investigates the thermomechanical behaviour of NiTi under such low-amplitude cyclic loading. This is carried out by analysing the frequency domain content of temperature sampled over a time window. The amplitude and phase of the most significant harmonics are obtained and compared with the theoretical predictions from the first and second-order theories of the Thermoelastic Effect, this being the typical reversible thermomechanical coupling prevailing under elastic straining. A thin strip of NiTi, exhibiting a fully superelastic behaviour at room temperature, was investigated under low-stress amplitude tensile fatigue cycling. Full-field strain and temperature distributions were obtained by means of Digital Image Correlation and IR Thermography. The work shows that the full field maps of amplitude and phase of the first three significant temperature harmonics carry out many qualitative information about the stress and structural state of the material. It is, though, found that the second-order theory of the Thermoelastic Effect is not fully capable of justifying some of the features of the harmonic response, and further work on the specific nature of thermomechanical heat sources is required for a more quantitative interpretation.