Identifying B19 crystal structure

Dear LAMMPS users,

I am currently running a tensile test simulation as part of my MSc thesis and I want to see the changes in phase that are taking place during the deformation of Cu-Zr binary alloy. It has already been reported experimentally that Cu-Zr undergoes a stress induced phase transformation from B2 to B19’ crystal structure (so there is a austenite to martensite stress-induced phase transformation).

I want to confirm this in my simulation. I am currently using Ovito, which uses common neighbour analysis to identify crystal structures but it only determines whether the crystal structure is bcc, fcc, hcp or icosahedral. In few words, it cannot identify a B19’ crystal structure (martensite) as far as I know.

Does anyone know how to indetify the B19’ crystal structure using molecular dynamics? I will appreciate your help.

Kind Regards,

Fabian

I am thinking that Ovito probably uses some sort of heuristic to identify the crystal structure. For example, for an FCC you know what the ideal distances to the nearest neighbours within a certain distance from the reference atom are, and you also know what the angles between the bonds should be. Therefore, if the actual observed distances and angles do not deviate too far from the reference, it is probably safe to assume that you have the corresponding phase.

The easiest thing to do would be to come up with a similar suitable criterion for the B2 and B19’ crystal structures, and use that to decide whether the crystal is in the one or the other phase. I have no idea what a B2 and B19’ phase look like so I cannot help much there, but I have used a similar approach in the past for determining whether or not a system had a certain symmetry (see the SI of this: http://pubs.rsc.org/en/content/articlelanding/2016/sm/c6sm00489j#!divAbstract). You just need to find a suitable way of discriminating the phases, which, depending on the problem, can be very easy or rather tricky.

There are different ways to look at this. One would to look at your rdf and compare it to an ideal rdf of B19’ (D Mutter, P Nielaba - Physical Review B, 2010 - APS). Also, you can compare the angle and lattice parameter of your B19’ to the experimental/DFT values.

Aria