Issue with sqs2tdb -fit in ATAT for Complex Lattices

Alloy Theoretic Automated Toolkit (ATAT)
1

Dear developers:

Thank you for your very helpful code and support.

I am currently using the sqs2tdb module in ATAT and encountered a question while performing the sqs2tdb -fit step.

Specifically, I tried fitting the TDB using both total energies and formation energies. For the FCC system in the Al-Mg-Si-Ag system, both approaches yield consistent results and generate identical TDB files. However, when I apply the same fitting process to a complex lattice, I obtain different TDB results depending on whether I use total energy or formation energy. I also noticed that the TDB file fitted using formation energy does not include the Gibbs free energies of the end members.

Do you have any suggestions on how to resolve this issue?
Best,

2

I’ve resolved the issue — it turns out that the first line in the terms.in file needs to be different for complex lattices with two sublattices compared to FCC lattices.
For example, the first line in the terms.in file for FCC lattice should be 1,0, while it should be 1,0:1,0 for complex lattices with two sublattices.

3

Yes. You need that for multi-sublattice systems. (And also for higher orders)

One example would be: Computational Assessment of Novel Predicted Compounds in Ni-Re Alloy System | Journal of Phase Equilibria and Diffusion