ABINIT Discussion Forums

Hi, I am a newbie to abinit and would like to ask regarding LSDA options.
The unitcell (vo2) is paramagnetic and I am trying to use LSDA, different spin up/down density.
Spin polarization is active using nsppol=2 but is there any other option in which I can employ the difference of spin up/down?
I mean "Nel_up - Nel_down" like "nupdown" in VASP code.

I read through the tutorial files but available ones are:

1. manual occ. input - will work on simple structure.
2. magnetic moment - FM or AFM. But my system is not FM or AFM either.

I am trying to use d-electrons of V as polarized one. Any comment will be very appreciated.

Byoungseon

http://www.abinit.org/documentation/hel ... tml#fixmom

will fix the relative chemical potentials of spin up and spin down so the occupations will give the desired total magnetization in collinear spin cases. This sounds like what you are asking about. Note that this is still a "ferromagnetic" case. True paramagnetism is something else - in DFT we usually just use non-magnetic simulations for paramagnetic cases.

Matthieu