search

mailhexu / TB2J

a python package for computing magnetic interaction parameters
BSD 2-Clause "Simplified" License
67 stars 29 forks source link

Error in lattice parameters #51

Closed lyonworton closed 3 months ago

lyonworton commented 3 months ago

I would like to ask why the lattice structure in the exchange.out file obtained from TB2J calculations is so much different from the crystal structure in my input file, causing the interface software to not find the correct symmetry when detecting the exchange.out file. Does this mean that the exchange parameters are calculated with such a structure and is the result reasonable?

mailhexu commented 3 months ago

Hi, Could you send me your input files , the exchange.out file, and the command you used to run TB2J? If you use Wannier90, could you send the .win files too? Thanks. The exchange parameters are computed with the Hamiltonian read from the DFT/Wannier code, which should be consistent with the DFT input files. But the distance calculated . TB2J use the ASE code to parse the structure files. Usually it can correctly parse the DFT inputs. But not all the DFT input syntax are recognized. HeXu

lyonworton commented 3 months ago

2DFe3GeTe2.zip Thanks for your reply! I'm using abacus to calculate the task, and the input structures of exchange.out and STRU are different, and when I'm reading the exchange.out file with the interface software to calculate the magnon band structure, I can't get the correct symmetry.

mailhexu commented 3 months ago

Hi, Thanks for the files! I don't see the difference between the STRU file and the exchange.out file. The only different I see is that in STRU file the positions are in reduced coordinates, and in exchange.out it is in cartisian coordinates. But perhaps I missed something? Could you send what discrepancy you notice? Which interface software did you use to compute the magnon band structure? If you use TB2J_magnon, I can see that the automatic k-path is the hexagonal one. This is not what you expected? Thanks a lot! HeXu

mailhexu commented 3 months ago

BTW, I noticed that the magnon band structure is computed before you did the "merge and rotate" procedure, in which you need to compute the exchange parameters with spin along x/y/z and merge. https://tb2j.readthedocs.io/en/latest/src/rotate_and_merge.html With Abacus, you don't need to rotate the structure as putting spin along different directions is possible. HeXu

lyonworton commented 3 months ago

Thank you for your reply, I am very sorry for the misunderstanding and the delay of your precious time. I mean the lattice-vectors are different, the precision is around 10-5, and the interface software I use is rad-tools, this software shows that the symmetry is not detected by reading the exchange.out file, and it is not the structure of HEX, so I guess it's because the CELL (Angstrom) parameter in exchange.out doesn't match the precision of the input file. In addition, thank you very much for your suggestion I will study it carefully. 图片1

mailhexu commented 3 months ago

My guess is that in the stru file, you put the unit Angstrom/Bohr as 1.889726 which makes the difference for the conversion. Could you try: 1.8897261258369282

But I don't think this will change the symmetry detection, as everything is scaled.

HeXu

On Wed, May 22, 2024 at 10:53 AM lyonworton @.***> wrote:

Thank you for your reply, I am very sorry for the misunderstanding and the delay of your precious time. I mean the lattice-vectors are different, the precision is around 10-5, and the interface software I use is rad-tools, this software shows that the symmetry is not detected by reading the exchange.out file, and it is not the structure of HEX, so I guess it's because the CELL (Angstrom) parameter in exchange.out doesn't match the precision of the input file. In addition, thank you very much for your suggestion I will study it carefully. 1.png (view on web) https://github.com/mailhexu/TB2J/assets/112809823/73aa97f8-5d38-42e6-a612-a38b99dc2970

— Reply to this email directly, view it on GitHub https://github.com/mailhexu/TB2J/issues/51#issuecomment-2124230481, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABUNK3EAPF4GMQBK3B75XI3ZDRMJZAVCNFSM6AAAAABIDBQKQCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRUGIZTANBYGE . You are receiving this because you commented.Message ID: @.***>

lyonworton commented 3 months ago

Ok, thanks for your patience in replying, it really helped me a lot, thanks a lot!