magnetic moment optimization

Message

mssb2 · #1 Post by **mssb2** » Fri Oct 27, 2006 9:46 am

Hi,
I am doing cluster calculation consisting of Transition metal atom & Silicon atom.
I did spin polarize calculation. I got a value of magnetic moment.
But I am not sure wheather the moments are optimized or not. The magnetic moment of my system mainly comes from the TM. To optimize th moment I mean to find out the the spin configuration of the TM for which my system will have minimum energy.
Is it possible in VASP ? What should I do for this?

Thanking You.

#2 Post by **admin** » Tue Nov 14, 2006 12:36 pm

the magnetic moments are determined self-consistently together with the charge densities, as the difference between spin up and spin down charge. I am not quite sure what you mean by the "spin configuration":
-- if you just do a collinear calculation, the total moments are given
-- the orientation of the spins (with respect to the quantisation axis, which is z unless set explicitely) can be obtained by using the LNONCOLLINEAR = True option. Please have a look in the online handbook on how to give the starting guesses (MAGMOM) and the quantisation axis (SAXIS) in INCAR then

bandy · #3 Post by **bandy** » Wed Nov 15, 2006 1:38 pm

The difference between spin up and spin down charge (expressed in unit 'e') should be integers: any explanation for a value say 2.71 or 0.58 e.

Thanks