Search

The martensitic transformation serves as the basis for applications of shape memory alloys (SMAs). The ability to make rapid and accurate predictions of the transformation temperature of SMAs is therefore of much practical importance. In this study, we demonstrate that a statistical learning approach using three features or material descriptors related to the chemical bonding and atomic radii of the elements in the alloys, provides a means to predict transformation temperatures. Together with an adaptive design framework, we show that iteratively learning and improving the statistical model can accelerate the search for SMAs with targeted transformation temperatures. The possible mechanisms underlying the dependence of the transformation temperature on these features is discussed based on a Landau-type phenomenological model. [ABSTRACT FROM AUTHOR]

Abstract: During repeatedly imposed thermally induced martensitic transformations in Ti–Ni shape memory alloys, the martensite start temperature Ms decreases. This has been rationalized on the basis of a scenario where an increasing dislocation density makes it more and more difficult for martensite to form. However, it is not clear why dislocations which form because they accommodate the growth of martensite during the first cooling cycle should act as obstacles during subsequent transformation cycles. In the present work we use diffraction contrast transmission electron microscopy to monitor the formation of unique leaf-like dislocation substructures which form as the martensite start temperature decreases during thermal cycling. We interpret our microstructural results on the basis of a microstructural scenario where dislocations play different roles with respect to the propagation of a big martensite needle in one transformation cycle and the nucleation and growth of new martensite needles in the following cycles. As a consequence, chestnut-leaf-like dislocation arrays spread out in different crystallographic directions. [Copyright &y& Elsevier]