Decompose the exchange into orbital contributions.

[Rick Anteur]

Hi,

I would like to know, if it’s possible to specify the orbitals to be considered in the decompostion; when we use tb2j (by wannier functions (for DFT QE )). 

[Xu He]

Yes. See https://tb2j.readthedocs.io/en/latest/src/orbital_contribution.html

Best wishes,

HeXu

On 9/3/21 5:06 PM, Rick Anteur wrote:

Hi,

I would like to know, if it’s possible to specify the orbitals to be considered in the decompostion; when we use tb2j (by wannier functions (for DFT QE )). 

--
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/92c5a338-5774-478a-9c74-866b13a21088n%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Rick Anteur]

Hi,

I ran wann2J.py --elements Mn_3d .... After some time, I got an error message (see attached file). However, when I run wann2J.py --elements Mn ...., the calculation works fine and all orbitals have been considered.

[何旭]

Hi,

Thanks!! You've found an mistake in the documentation. 

This "_3d" option is for the SIESTA backend only (as the information which orbital is 3d is not present in the Wannier output, therefore it is not possible to select). I have made the changes in the documentation. 

Best regards,

HeXu

[Rick Anteur]

Hi,

Thanks a lot for your answers!

Here is the result I get (see file). The band structures opted by DFT agree well with those obtained with wannier90. Do you have an idea of ​​the choice of all its orbitals by tb2j? Whereas in my input file wannier only the d orbitals for the projections; have been chosen (.wout also only understands d orbs).

I also added when running tb2j, the parameter '' --exclude_orbs '' to choose 5 orb for all the atoms of Mn. But it seems that the orbitals that I exclude, are still taken into account. There is not too much information on its use (- exclude_orbs). Can you tell me more about its use?

Thanks a lot!

[Rick Anteur]

sorry !

[何旭]

Hi,

Thanks for the data. 

The way TB2J choose the orbitals for one atom is that, each Wannier function is assigned to the atom closest to it.

 From the data it seems the Wannier functions are not well converged. Did you check the Wannier centers and the spread in the .wout file to see they are close to the atoms and the spread are small? A good band structure does not guarantee the good Wannier functions.

I will check if the --exclude_orbs works or not. 

Best regards,

HeXu

[何旭]

Hi, 

About the --exclude_orbs option, could you check the latest version (0.6.6). I tested it and it works. 

Best regards,

HeXu

[Rick Anteur]

Hi,

About the —exclude_orbs option, you said you tested it in version 0.0.6 and it works.  Can you share the command to type for the -exclude orb computation?

Best regards,

[Xu He]

Hi,

Here is one example.

In a Wannier input file, the s+p+d orbitals are included for an Fe atom:

begin projections
Fe:s;p;d
end projections

If only the contributions from the d orbitals are needed, put

--exclude_orbs 0 1 2 3

in the command (note that python counts from 0).

to exclude the s and p orbitals.

Best wishes,

HeXu

--
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/CAMRadMUj_YrJnnf8ihQZo94PoEmWqBtDF0Nq7%2BsvJg2RSnFd2A%40mail.gmail.com.

[Rick Anteur]

Hi,

Awesome!  Thank you so much ?

Best regards!