My Community

Hi,

I have Oxygen (and also OH) on a graphite sheet (graphene).
With the Oxygens, I suppose I should do a spin-polalized calculation.

From the guidebook I learned that I first should do calculations
without spin-polarization: relax and accurate total energy.

From that last result, I keep the CHGCAR and WAVECAR files
and calculate accurate total energy with ISPIN=2 and ICHARG=1.
From the Handson lectures (from VASP webpage),
I learned that I also should set reasonable values for MAGMOM.

My main question is, how to set reasonable values for MAGMOM in
this graphite - Oxygen - OH system. Say, I have in my supercell
8 C-atoms, 2 Oxygen and 1 Hydrogen, would this then be a good
start:
MAGMOM = 8*0 2*2 1*0
assuming that the Carbons and the Hydrogen will not contribute to the
magnetic moment, but the Oxygens will with a magnetic moment of 2
each. Is that a reasonable assumption? Or would this be better:
MAGMOM = 8*2 2*2 1*1


My POTCAR file has "LEXCH = PE"; should I then NOT use VOSKOWN,
as that is only desirable with PW91 functionals

Thank you,
Rob.

if you assume that any atom will not have a significant spin-polarization, you can safely start by setting MAGMOM=0 for that atom. The self-consistency procedure will then stabilize the scf solution of the local moments as well.

VOSKOWN only has to be set for PW91.