How to visualize the band decomposed density for beta channels
Posted: Tue Dec 04, 2012 1:57 am
Greetings VASP community,
I have a question in regard to the title. I have completed a spin-polarized HSE06 calculation on a large supercell of a magnetic system.
To achieve magnetic solution, I use these flags (for example):
NUPDOWN=4
MAGMOM= 2 2 0 0
My magnetic moments are reasonable in the converged job (slightly less than +2 e- per paramagnetic atom {~1.8 e-}.)
I then run a band-decomposed charge density to obtain the eigenstates for visualization. However, this approach only images the alpha eigenvalues.
To image the beta channels, I change only flags related to the magnetization like this, and run a new job entirely:
NUPDOWN=-4
MAGMOM=-2 -2 0 0
This also converges, but the spins on the atoms are greater than -2 (say -.9). This is, presumably, due to the fact that NUPDOWN is a NEGATIVE number, allowing a relaxation (per the manual). I cannot simply flip the sign of the magnetization, since NUPDOWN is in reference to the alpha channel only (4 more in alpha than beta).
The goal of this business is to "flip" the former beta channel into the new alpha channel, and visualize this way, circumventing VASP limitation (
Does anyone have a better suggestion for how to obtain the band decomposed density for a beta channel for a SP job?
Thank you.
<span class='smallblacktext'>[ Edited ]</span>
I have a question in regard to the title. I have completed a spin-polarized HSE06 calculation on a large supercell of a magnetic system.
To achieve magnetic solution, I use these flags (for example):
NUPDOWN=4
MAGMOM= 2 2 0 0
My magnetic moments are reasonable in the converged job (slightly less than +2 e- per paramagnetic atom {~1.8 e-}.)
I then run a band-decomposed charge density to obtain the eigenstates for visualization. However, this approach only images the alpha eigenvalues.
To image the beta channels, I change only flags related to the magnetization like this, and run a new job entirely:
NUPDOWN=-4
MAGMOM=-2 -2 0 0
This also converges, but the spins on the atoms are greater than -2 (say -.9). This is, presumably, due to the fact that NUPDOWN is a NEGATIVE number, allowing a relaxation (per the manual). I cannot simply flip the sign of the magnetization, since NUPDOWN is in reference to the alpha channel only (4 more in alpha than beta).
The goal of this business is to "flip" the former beta channel into the new alpha channel, and visualize this way, circumventing VASP limitation (
Does anyone have a better suggestion for how to obtain the band decomposed density for a beta channel for a SP job?
Thank you.
<span class='smallblacktext'>[ Edited ]</span>