[Wien] How to set a specific spin state

[Peter Blaha]

Hi,

We do "self-consistent" calculations. This means that in most cases, the 
electronic configuration of the atoms does not matter.

Only in some magnetic cases, the results may depend on the starting 
configuration, and you may be able to obtain a high-spin and low-spin 
solution, depending on the starting density (The same happens with the 
magnetic order, i.e. ferro-magnetic vs. different anti-ferromagnetic 
ordering).

Thus, all we specify in case.inst is the number of spin-up and spin-dn 
electrons, but we do NOT select a particular d-orbital. Remember, lstart 
produces anyway only spherical densities.

So for the low-spin state of Co put the same number of up and dn 
electrons into case.inst, for the high-spin state, start with a large 
atomic moment.
The specific occupation of t2g or eg will come out automatically in the 
scf prodedure.

Second question:   lstart was a non-spinpolarized code (and the input is 
still in a non-spinpolarized way, i.e. it is given in a relativistic 
notation, but the spin-occupation is defined without considering the 
spin-orbit coupling). Thus you cannot calculate a "proper" spin-state in 
lstart (with spin-orbit coupling).
All what lstart gives is a spherical !! spin-density, which gives the 
desired spin moment when integrated over space.

This is all we need as an input for the scf cycle.

PS: With PBE+U you can manipulate the density matrix and then force 
eventually the occupation of certain d-orbitals. (There should be 
previous postings on that).

On 9/28/20 9:47 AM, Yfding wrote:
> Dear Wien2k developers，
> 
> In Wien2k, case.inst file allows to define the spin-state of certain 
> atoms, and it consists of the relativistic quantum number. It seems that 
> relativistic quantum number cannot be directly connected with the 
> occupation of 3dx2-y2, 3dz2, 3dxy, 3dxz, 3dyz. My question is, how do I 
> set a specific spin state, like low spin Co3+ ( t2g orbit is filled with 
> 3 spin up electrons and 3 spin dn electrons to make the total spin 
> equals zero while no electron exist in eg orbit ), by rewriting 
> case.inst file ?
> 
> The second question is about Table 6.6: Relativistic quantum numbers in 
> user's guide. In the last two columns, max. occupation of s = -1 is 4, 
> and max. occupation of s = +1 is 6 for l = 2. It seems that max. 
> occupation for d orbit will lead to total spin equals (-1)*4+1*6=2. Is 
> it right ?
> 
> Any comments would be highly appreciated. Thanks in advance!
> 
> _______________________________________________
> 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
> 

-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at/TC_Blaha
--------------------------------------------------------------------------