I'm trying to do a spin-orbit DOS calculation on UAu2, and the interstitial DOS is symmetric in the up and down spins, but shows their density of states swapping sign twice (so one spin species starts as positive then smoothly crosses to negative and back again later, while the other species mirrors this). Neither the total nor the partial plots swap sign in this way.
I thought this might be an issue with muffin tins overlapping, but this isn't the case after doing some checks. Anyone know of why this might be happening? And if any of the DOS can be trusted?
Thanks!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
your information is very sparse and it is not clear what results you actually expect?
I am now guessing:
Do you run a AF structure with SOC included?
Do you have a spin polarized DOS on all sites as well as total?
If so, your integrated difference in DOS (up minus down) gives you the z-component of the local or total magnetization. On the atomic sites it probably is non-zero but in intestitial almost zero, right?
Then it makes perfect sense and the DOS can be trusted if you have made no errors ...
/Lars
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Apologies for being cryptic, attached is a plot of the interstitial dos where you can see the two lines crossing over.
The partial dos for each atom is symmetric in both spin species, and the total dos is too (with neither of them changing sign like the interstitial).
Hopefully that's more clear.
Hi all,
I'm trying to do a spin-orbit DOS calculation on UAu2, and the interstitial DOS is symmetric in the up and down spins, but shows their density of states swapping sign twice (so one spin species starts as positive then smoothly crosses to negative and back again later, while the other species mirrors this). Neither the total nor the partial plots swap sign in this way.
I thought this might be an issue with muffin tins overlapping, but this isn't the case after doing some checks. Anyone know of why this might be happening? And if any of the DOS can be trusted?
Thanks!
Dear Harry,
your information is very sparse and it is not clear what results you actually expect?
I am now guessing:
Do you run a AF structure with SOC included?
Do you have a spin polarized DOS on all sites as well as total?
If so, your integrated difference in DOS (up minus down) gives you the z-component of the local or total magnetization. On the atomic sites it probably is non-zero but in intestitial almost zero, right?
Then it makes perfect sense and the DOS can be trusted if you have made no errors ...
/Lars
Hi Lars,
Apologies for being cryptic, attached is a plot of the interstitial dos where you can see the two lines crossing over.
The partial dos for each atom is symmetric in both spin species, and the total dos is too (with neither of them changing sign like the interstitial).
Hopefully that's more clear.
Harry
Last edit: Harry K 2017-02-01