[Wien] Can we perform layered AFM calculations with Wien2k?

[pieper]

Hello Krishnaveni Parthasarathy

Yes you can - though admittedly I did not look up your reference, and I 
might misinterprete your idea of 'layered AFM'. And I suppose you 
consider collinear magnetic moments. If that is not the case you might 
look into the (unsupported) WienNCM.

In general, study the UG, especially the chapters on AFM calculations.

You can think of any collinear AFM as ferromagnetic layers stacked along 
one particular crystalographic direction, the direction to the nearest 
antiparallel neighbor. For calculations with Wien2k you need a .strcut 
with a unit cell that has at least two magnetic atoms along that 
direction (perhaps from supercell if the crystalografic primitve cell is 
not large enough).

Since you want them to be magnetically inequivalent the two have to be 
on inequievalent crystalografic sites. If they are on crystalografic 
equivalent positions you have to split the position and number the atoms 
to tell Wien2k that they are not same (perhaps Mn1 and Mn2). Note that 
this results in a different symmetry of your structure.

If you want to compare total energies for different magnetic structures 
be careful to calculate all of them on the same crystalographic unit 
cell, changing only in the initialization which magnetic atom belongs to 
the group pointing 'up' and which one to the group pointing 'down'. If 
you have, for example, constructed a supercell with four inequivalent 
Mn-sites Mn1 .. Mn4, you can calculate AFM structures corresponding to 
Mn1, Mn2 up and Mn3, Mn4 down, or Mn1,3 up and Mn2,4 down, or ... This 
will correspond to ferromagnetic layers of Mn1, Mn2 stacked with 
antiparallel ferromagnetic layers Mn3,4, or ferromagnetic layers of 
Mn1,3 ... Not all of these arrangements are necessarily different. If 
the direction Mn1 - Mn2 is symmetry related to the one Mn1-Mn3 in the 
underlying crystalografic lattice, then there is no need to calculate 
with the stacking in these two equivalent directions. And obviously, for 
a complicated AF wave vector (the direction of the stacking), you will 
need large supercells so think, before you start.

Good luck,

Martin Pieper


---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564


Am 28.09.2015 18:43, schrieb Krishnaveni. S:
> Ref: Electronic Structure, Chemical Bonding, and Finite-Temperature
> Magnetic Properties of Full Heusler Alloys
> 
> Authors: YASEMIN KURTULUS, MICHAEL GILLEßEN, RICHARD DRONSKOWSKI*
> 
> DOI 10.1002/jcc.20308
> 
> online in Wiley InterScience (www.interscience.wiley.com [1]).
> 
> Ref to the article (Page 93) mentioned above, I understand that one
> can perform layered calculation in AFM calculations.
> 
> In page 93, the authors have mentioned as below: “Here, the symmetry
> has been slightly reduced to obtain a model in which the Mn planes can
> be treated separately. This particular alignment is characterized by
> alternating planes along (001) of spin-up and spin-down manganese
> moments, and the identically oriented manganese moments within each
> plane are coupled by the p orbitals of the Z elements. An alternative
> anti ferromagnetic model with alternating magnetic planes along (111)
> is also thinkable, but earlier total-energycalculations11 have already
> clarified that the (001) model is characterized by lower energies in
> almost all cases.”
> 
>  The author has specified that these calculations have been done in
> VASP. Can some one help me understand if it is possible to perform
> similar type of calculations in Wien2k?
> 
> --
> 
> Thanks and regards
> 
> Krishnaveni Parthasarathy
> 8939675012
> 
> Links:
> ------
> [1] http://www.interscience.wiley.com
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html