about dmi

[Jie Kang]

Dear HeXu,

Recently, when using VASP+wannier+TB2J to calculate the DMI of BiFeO3, I found that the DMI vector of the first nearest neighbor has a significant difference in data, but the J is almost the same value. What is the reason for this? How should I solve the problem of differences in the nearest neighbor DMI vector data? My calculation results have been uploaded in the attachment.

The second question is, I found that the parameter num_iter has a significant impact on the calculation of the exchange constant. When I set it to 0, I can already get a good fitting of the Wannier function. Why is it necessary to localize 300 steps in the example? Can you tell me when localization is needed? It seems that excessive localization will break the symmetry of the structure.

Best regards,

Rong Chen

[matthieu verstraete]

On Tue, Apr 16, 2024 at 4:04 PM Jie Kang <fddm20...@gmail.com> wrote:

Dear HeXu,

Recently, when using VASP+wannier+TB2J to calculate the DMI of BiFeO3, I found that the DMI vector of the first nearest neighbor has a significant difference in data,

difference between what? If it's that the vectors for symmetry equivalent neighbors are different, indeed DMI is much more sensitive (and usually much smaller) than J. My recommendation at this stage would be to average and symmetrize the different vectors and rotate them back to their individual positions.

 

but the J is almost the same value. What is the reason for this? How should I solve the problem of differences in the nearest neighbor DMI vector data? My calculation results have been uploaded in the attachment.

 

The second question is, I found that the parameter num_iter has a significant impact on the calculation of the exchange constant. When I set it to 0, I can already get a good fitting of the Wannier function. Why is it necessary to localize 300 steps in the example? Can you tell me when localization is needed? It seems that excessive localization will break the symmetry of the structure.

yes - you are correct, this is a known effect, and using a (less localized) well centered and relatively symmetric atomic orbital projected WF from 0 steps is better. We are working on other optimization and centering schemes. Stay tuned!

Best

Matthieu

 

Best regards,

Rong Chen

--
You received this message because you are subscribed to the Google Groups "TB2J" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tb2j+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tb2j/f534a60d-fa79-407e-8ea6-39a543d1ed9fn%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Professor Matthieu J Verstraete

Fellow, American Physical Society

Chair, Steering Committee, European Theoretical Spectroscopy Facility www.etsf.eu

Alumnus Fellow, Young Academy of Europe  yacadeuro.org

Institute for Theoretical Physics (ITP) Department of Physics

Buys Ballot Gebouw/Building, Princetonplein 5, office 

University of Utrecht,  3584 CC Utrecht

ITP Secretariat: +31 30 253 5928 E-mail: science....@uu.nl

Group web page (Liège): http://www.nanomat.ulg.ac.be/

Nanomat lab, Q-Mat center, Université de Liège
Département de Physique, Bat. B5a, 4/50
Allée du 6 août, 19 B-4000 Sart Tilman, Liège Belgium
Phone : +32 4 366 90 17

European Theoretical Spectroscopy Facility (ETSF)

Mail : matthieu....@uliege.be
          matthieu.jea...@gmail.com

          m.j.ver...@uu.nl

[Virginie Muzala]

Please i don't speak englih

To view this discussion on the web visit https://groups.google.com/d/msgid/tb2j/CAB--OUxtr5msJhRyd8XLjWeqDywhg21xb%3DLECiR1j9JDgT_Oww%40mail.gmail.com.

[Jie Kang]

Dear  Matthieu,

Thank you for your reply. I have the following questions.

1.I want to know the reason why the DMI vector values of the nearest neighbor atoms are different? Is it because the crystal structure is an imperfect crystal structure? Can we still take the average if the values of the nearest neighbor DMI vectors differ too much? What does it mean to rotate their back to their individual positions? How to operate specifically? Can I directly apply magnetization in three different directions without rotating the lattice to obtain DMI?

2.How to determine the value of num_iter and obtain reliable results?

Best regards,

Rong Chen

[matthieu verstraete]

On Thu, Apr 18, 2024 at 3:36 AM Jie Kang <fddm20...@gmail.com> wrote:

Dear  Matthieu,

Thank you for your reply. I have the following questions.

1.I want to know the reason why the DMI vector values of the nearest neighbor atoms are different? Is it because the crystal structure is an imperfect crystal structure?

your crystal structure gives you your real symmetries, then the WF break them - they are the main problem in the different DMI vectors.

 

Can we still take the average if the values of the nearest neighbor DMI vectors differ too much?

depends, it's always an approximation.

 

What does it mean to rotate their back to their individual positions?

the DMI vectors should rotate one into the other with symops. 

 

How to operate specifically? Can I directly apply magnetization in three different directions without rotating the lattice to obtain DMI?

2.How to determine the value of num_iter and obtain reliable results?

As mentioned, 0 is probably the safest right now.

 

To view this discussion on the web visit https://groups.google.com/d/msgid/tb2j/93538fd2-b414-4569-a1c7-d52fc1232da7n%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.