ABINIT Discussion Forums

Hi all,

I have a compound with four elements, let's call them A, B, C and D, I'm interested in making elements A and B disordered in the lattice, and elements C and D ordered. This just means that elements C and D always occupy the same coordinates, while the coordinates of A and B can be randomized in the compound. I found an input variable 'random_atpos' that can potentially be used for this purpose, but does anyone know if it's possible to have just elements A and B disordered, while keeping C and D on ordered lattice coordinates?

Thanks a lot.

Hi

I don't know if abinit can do that. Alternatively, you can use a special quasi random structure (SQS) as input structure for abinit. The SQS will be a small supercell that will mimic as closely as possible the random distribution of your A and B elements. You can use ATAT (http://www.its.caltech.edu/~avdw/atat/) to build the SQS (see the manual). It usually works well for modelling alloys.

Good luck

Boris

Ok thank you very much for your help Boris.