Magnetic anisotropy splitting over atoms

[rc1ph]

I calculate magnetic anisotropy (MA) as total energy difference between 100 and 001 spin directions. Generally no problems following VASP manual.

To understand the source of magnetic anisotropy, it would be very helpful to know the corresponding contributions from each atom in the system to the total magnetic anisotropy.

In OUTCAR, seems there is a subdivision "Ion # E_soc:" of the total spin-orbit coupling energy into separate atoms. If it is true, then for each atom_i:

MA(atom_i)=E_soc(atom_i,100)-E_soc(atom_i,001).

The results look reasonable as function of atom_i.

But if I try to sum up E_soc over all atoms and then make the difference between 100 and 001:

Sum_i [E_soc(atom_i,100)] - Sum_i [E_soc(atom_i,001)],

I should get the same total anisotropy of system as obtained from total energy difference between 100 and 001 spin directions.

Unfortunately, the results may be different by 1.5x.

I tried to vary RWIGS, but seems E_soc do not depend on RWIGS.

The layered system is Fe5|(MgO)3.

So,
1. Are E_soc really the subdivision of spin-orbit coupling energy into separate atoms?

2. Why do they not depend on RWIGS and how such subdivision over atoms is performed?

3. Why the summation of E_soc over atoms does not reproduce the total anisotropy obtained from total energy difference?

The answer would be greatly appreciated as being crucial for our current research.
Thanks,
Roman

[support_vasp]

Hi,

We're sorry that we didn’t answer your question. This does not live up to the quality of support that we aim to provide. The team has since expanded. If we can still help with your problem, please ask again in a new post, linking to this one, and we will answer as quickly as possible.

Best wishes,

VASP