How Do Martensitic Twin Boundaries Move?

  • PDF / 267,365 Bytes
  • 6 Pages / 612 x 792 pts (letter) Page_size
  • 44 Downloads / 198 Views

DOWNLOAD

REPORT


How Do Martensitic Twin Boundaries Move? Graeme J. Ackland and Udomsilp Pinsook1 Department of Physics and Astronomy, University of Edinburgh, Edinburgh EH9 2LZ, UK 1 Department of Physics, Chulalongkorn University, Bangkok, 10330, Thailand. ABSTRACT The martensitic phase transformation from bcc to hcp underlies a number of curious effects, including shape-memory and superelasticity. The distinctive feature is the microscopic reversibility of the transition: at the atomic level, the position of each atom is uniquely related between phases. Moreover, this one-to-one relationship holds not only for the perfect crystal transformation mechanism, but also for the topological defects (such as twin boundaries) which are created in the transformation. Furthermore, for shape memory effect the microscopic relationship must be preserved by the deformation mechanism active in the material. In this paper, we examine the microscopic phenomena which allow these relationships to hold, and the consequences for shape-memory alloy design. INTRODUCTION We wish to investigate the atomic-level processes involved in twin boundary motion in shapememory materials. We start by considering the Nishiyama-Wassermann mechanism for the bcc-hcp phase transition in an elemental material. This may appear strange, because in practice, real shape memory materials are alloys and do not have the bcc crystal structure. However, the high temperature phase of shape memory alloys typically has the binary or ternary equivalent of the bcc phase, and the martensite is typically similar to the hcp structure. Understanding the simpler geometry of the elemental phase facilitates discussion of the alloy structures. The results presented here are a summary of geometric, molecular dynamic and electronic structure calculation. We will concentrate here on the emergent structural and geometric aspects: the calculational details can be found elsewhere. The results are divided into two sections; the first covers the detailed mechanism by which martensitic twin boundaries move in the hcp structure. We then discuss generalisations of the mechanism to the binary alloy case, in particular the B2-B19’ phase transformation in NiTi, and describe which twin boundaries satisfy the microscopic reversibility requirement for shape memory effect. CALCULATIONS AND RESULTS In the Nishiyama-Wassermann transition path, one set of (011) planes of the bcc structure is transformed into close packed planes of the hcp structure. There is also a shuffle of alternate planes, corresponding to the T1N phonon in bcc. This transformation will occur martensitically only for materials where the bcc structure is not a minimum of the total energy, i.e. the interatomic interactions are such that bcc is unstable (not metastable) at zero temperature. A schematic contour plot of the energy of such a structure is shown in figure 1, alongside the phonon dispersion relation for zirconium calculated using a many-body potential. In this figure, the instantaneous structure of a region of the crystal is represent