[Wien] Help on AFM Structure

[Gavin Abo]

> Anyway, when you have a structure which has only a "few" (2-4) 
> nonequivalent Fe positions, then it makes sense to play with magnetic 
> order and compare total energies.
>
> You would start in different directories and BEFORE init_lapw run 
> instgen_lapw -ask and specify the desired magnetic order by chosing up 
> or dn for each Fe site (and nm for Ga).

Slide 26 at the link below for another material YBaFe2O5 might be a good 
complementary example for this type of thing:

http://www.cms.tuwien.ac.at/media/uploads/cms/psi-presentations/Blaha.ppt

and corresponding article:

http://publik.tuwien.ac.at/files/PubDat_174458.pdf

>
> For visualisation you may create a struct file where you rename your 
> atoms to Co, Ni,...., so that you can distinguish them in xcrysden and 
> then you can see how to arrange the magnetic moments to get them in a 
> specific order.

Good to know, but one might also try VESTA.  In VESTA, you should be 
able to turn on atom labels individually without having to rename them 
in the struct file.  Also, you can hide the Ga atoms, if you just want 
to see the Fe atoms.