Relaxation convergence of the structures

Alloy Theoretic Automated Toolkit (ATAT)
1

Hello,

I started using ATAT interfaced with VASP recently to get the ground state structure of a pseudo-binary system. I have two questions.

  1. Out of all the structures generated, few relaxation calculations are not completely converged and ATAT is going ahead leaving those structures. So, If I have to re-run those unconverged calculations by copying CONTCAR (relaxed structure) to POSCAR or by reducing the convergence criteria, how do I do that and will maps consider those structures.

  2. During the calculation, some structures are automatically deleted by ATAT (for example, out 30 structures generated, only 12-13 are remaining in the folder). I would like to know how this operation of deleting certain structures is carried out.

Thank you.

Regards,
Shiva.