SQS for 4 equiatomic alloys

Alloy Theoretic Automated Toolkit (ATAT)
1

Hello everyone!
I am new to the ATAT Package, i want to know if we generate a fcc sqs for the composition (Fe 0.25, Ni 0.25, Cr 0.25, Co 0.25) for lets say 250 atoms so as to maintain equi-length sqs cell?
Does SQS takes into account entropic and enthalpic contributions for a particular configuration?
Is it possible to use best sqs structures in MD and replicate the box in some direction and minimize it?
Is SQS better than generating random structures using set type commands in lammps?
What should be the strategy to generate equilength SQSCell?

I know these may be trivial for folks to answer?
Thanks,
Mayank

2
  1. use a multiple of 4 instead of 250
  2. SQS account for enthalpic contributions, under the assumption of a perfectly disordered alloy. you have to add entropy yourself.
  3. You can "replicate the box in some direction" if you want, but I don’t know why you’d want to do that.
  4. For small cells (say <100), an sqs is significantly better, but the advantage decreases as cell size grows.
  5. you can specify you own cell. As explained in mcsqs -h
    The -rc option lets you specify supercells to use (via a sqscell.out file).
3

Thanks so much… for the reply…it do clear lots of doubt.

My question is how it accounts for enthalpic contributions? If i wants to read that; kindly suggest;

Also is it possible to quantify these contributions?

I want to use SQS in lammps; will it be better than random generation command in lammps?
better i mean to say in predicting disordered alloys properties?

Anyways thanks so much

4

How to get the enthalpic contributions??? You just input that SQS in any code that can compute energies…

An SQS is a more accurate model of perfect disorder than a cell generated at random.
But for large cells, the difference may not be significant.

A classic reference is of course https://doi.org/10.1103/PhysRevLett.65.353