Non-collinear Calulations

Message

fuzzylogic · #1 Post by **fuzzylogic** » Thu Oct 07, 2010 4:37 am

Hi, I am new to vasp and would like help on the following.
I hv started with non-collinear calculations and would like to know how can we vary the orientation of an atom in vasp, Assuming we hv two atoms how do we say fix one in z- direction and vary the other in random directions.

forsdan · #2 Post by **forsdan** » Thu Oct 07, 2010 9:14 am

You can impose constraints on the magnetic moments for specific atoms by the means of penalty contributions, while leaving the magnetic moments free to vary on other atoms. Please have a look at

http://cms.mpi.univie.ac.at/vasp/vasp/C ... ments.html

( If this now is what you meant with your question. )

Cheers,
/Dan
<span class='smallblacktext'>[ Edited Thu Oct 07 2010, 12:11PM ]</span>

fuzzylogic · #3 Post by **fuzzylogic** » Thu Oct 07, 2010 10:18 am

Thanks forsdan but my question means that while specifying MAGMOM in non-collinear calculations we specify 3 different orientations for each atom. How can i then specify a random orientation on an atom say an angle of 20 deg in x-direction or any other angle on a single run.

forsdan · #4 Post by **forsdan** » Thu Oct 07, 2010 11:15 am

The entry form of the MAGMOM tag for non-collinear calculations is given in the manual ( http://cms.mpi.univie.ac.at/vasp/vasp/L ... R_tag.html ), so you can just directly supply the three components for each atom.
However, please keep in mind that the values are only a initial guess and the magnetic moments will relax in order to reach the closest local energy minimum, thereby possibly changing their direction and magnitude. If you don't want this to happen, you can impose constraints on the magnetic moments according to the link I posted above, in order to restrict the directions and magnitudes to the initial guess.

<span class='smallblacktext'>[ Edited Thu Oct 07 2010, 01:16PM ]</span>