Hi everyone!
I am trying to do FM & AFM calculations in doped DMS system. I want to specify initial magnetic moment to the impurity only - Mn atoms. I have a 38 atom supercell of In17Mn2As19. I want to assign magnetic moment to only Mn.
How to do it?????
For FM:
Magmom = 17*0 19*0 1 1
For AFM:
Magmom = 17*0 19*0 1 -1 - is it correctly?
My POTCAR file ofcorse contain the Mn pseudopotential twice.
Please help
Margo
Ferromagnetic and Antiferromagnetic calculations
Moderators: Global Moderator, Moderator
-
goja25
- Newbie
- Posts: 12
- Joined: Fri Sep 28, 2007 12:40 pm
- License Nr.: 974
Ferromagnetic and Antiferromagnetic calculations
Last edited by goja25 on Tue Feb 17, 2009 3:19 pm, edited 1 time in total.
-
forsdan
- Sr. Member
- Posts: 339
- Joined: Mon Apr 24, 2006 9:07 am
- License Nr.: 173
- Location: Gothenburg, Sweden
Ferromagnetic and Antiferromagnetic calculations
I don't see why you need the Mn pseudopotential twice. You can still specify the initial magnetic moments to the two Mn atoms by using
MAGMOM = 17*0 19*0 1 1
or
MAGMOM = 17*0 19*0 1 -1
in the two respective cases so the Mn pesudopotentials doesn't have to appear twice in the POTCAR-file. However, do you know if the two different magnetizations are stable? It might otherwise happen that the two calculations converge to the same results. If you know that there are two stable solutions and you know the magnetization values it can sometimes be good to multiple the known magnetic moments on the atoms by a factor of 1.2-1.5 and use that as the initial magnetic moment, This can improve the convergence to magnetic state in question.
Cheers
/Dan
<span class='smallblacktext'>[ Edited Wed Feb 18 2009, 07:32AM ]</span>
MAGMOM = 17*0 19*0 1 1
or
MAGMOM = 17*0 19*0 1 -1
in the two respective cases so the Mn pesudopotentials doesn't have to appear twice in the POTCAR-file. However, do you know if the two different magnetizations are stable? It might otherwise happen that the two calculations converge to the same results. If you know that there are two stable solutions and you know the magnetization values it can sometimes be good to multiple the known magnetic moments on the atoms by a factor of 1.2-1.5 and use that as the initial magnetic moment, This can improve the convergence to magnetic state in question.
Cheers
/Dan
<span class='smallblacktext'>[ Edited Wed Feb 18 2009, 07:32AM ]</span>
Last edited by forsdan on Wed Feb 18, 2009 6:30 am, edited 1 time in total.