Spin with VASP

An example of visualising antiferromagnetic spin in FeO with VASP.

The input files used for VASP were:

POSCAR
comment
 1.0000000000000000
    0.000000     4.332000     4.332000
    2.166000     0.000000     2.166000
    2.166000     2.166000     0.000000
  Fe  O
  2  2
Direct
 0.000000   0.000000   0.000000
 0.500000   0.000000   0.000000
 0.250000   0.500000   0.500000
 0.750000   0.500000   0.500000
INCAR
ISPIN = 2
MAGMOM = 1.0 -1.0 0 0
encut = 500
KPOINTS
comment
0
Gamma
3 3 3
0 0 0

These should be combined with a POTCAR formed from an Fe and O pseudopotenial from the VASP distribution. Then VASP can be run, and the output analysed.

$ ./c2x -sv --null CHG
First FFT grid     72 36 36
spins=2   spinors=1
Found 3D data for Spin
  min=-5.44952 max=5.44952 sum=1.86429e-12 int=8.12102e-16  int|s|=7.15543
  (integral is e per cell for charge and spin densities)

The spin is contained in both the CHG and CHGCAR output files, and c2x can read either.

At this point the text from the corresponding Castep example can be repeated. The spin density is sharply concentrated close to the iron atoms, which are displayed as largish spheres, so one must set the isosurface value at a surprisingly small fraction of the peak values in the dataset. Here the isovalue is 0.06. Two unit cells are shown, with the repetition being in the y direction. As oxygen and iron are both shown as red, Display, Atomic Symbols has been turned on.

FeO and spin density

The alternating spins in the layers of Fe are clearly seen.

The net spin per cell in this antiferromagnetic state is, of course, zero, to which 8e-16μB is a good approximation. The integral of the modulus of the magnetic moment is 7.15μB per cell, or 3.58μB per Fe. This is within the range of published values - a 2016 JCTC paper quotes an experimental range of 3.3 to 4.2μB, and a calculated range of 3.4 to 4.3μB.