Suggestions Use of fitfc and felec

Alloy Theoretic Automated Toolkit (ATAT)
1

Dear Dr. Axel and ATAT users,

I wish to calculate the vibrational and electronic contribution

2

I accidentally submitted the POST without posting the whole question…Extremely sorry about that…

Hello ATAT users,
I wish to calculate the vibrational and electronic contribution to Cp for a system using the fitfc and felec codes.
I generated a 72 atom supercell (hexagonal which was subsequently relaxed (ISIF=3) in VASP. Then as the manual says to generate the perturbations which was done using fitfc code. The structures were re-relaxed with ISIF=4. Now, I have a few clarifications to seek.
a. Does svib.out give the Gibbs energy or the Helmholtz energy ? I guess it is Gibbs energy. Please correct me…
b. Now, one of the input files for felec is dos.out. Can you please inform where to get the input file from? I mean using which code etc.
c. Thirdly, how to get the vibrational contribution to Cp from these codes. And how to get the electronic contribution to Cp by felec.
Can some of you elaborate on this…

Thank you

Regards
Suddhasattwa Ghosh

3

First, let me make sure of this: you should not re-relax the perturbations! These should be static calculations!
About your questions:
a. svib gives entropy. fvib gives the Helmholtz free energy unless you specified the -P=… option in which case in gives the Gibbs
b. dos.out if generated automatically if you used runstruct_vasp. If not, you can always try runstruct_vasp -ex which should extract the info for felec to use from the vasp output. (Make a backup of the files first, just to be sure!)
c. S=-dG/dT and Cp/T=dS/dT all at constant P.

4

Dear Axel,

I need some small clarifications.

  1. As you said, "you should not re-relax the perturbations"…When we use fitfc command for the first time (after relaxing the original structure)
    fitfc -er=11.5 -ns=3 -ms=0.02 -dr=0.1
    It creates vol_* directories…Now as per the manual , you mention that "all degrees
    of freedom except volume are allowed to relax" for these structures. That means ISIF=4 in the INCAR file. Am I correct?

  2. When the fitfc command is again invoked and it creates the subsubdirectories; I understand that these have to be static calculations (IBRION=-1 and NSW=0). Am I correct?

Thank you
Regards

Suddhasattwa Ghosh

5

You are correct on both points.
The calculations in vol_?/ are to be relaxed
The calculations in vol_?/*/ are to be static.