[Wien] Question in optimizing the position of a ferromagnetic material

[Peter Blaha]

First of all:

I hope you have done KFe2As2 bulk first.
Since it contains Fe, it could be that it is magnetic (FM or AFM ? check 
literature) and one should use   runsp
Check out if a magnetic moment persists, this answers your question if 
you should use run or runsp

Why do you initialize with   RKMAX=9.5 ??? This is not necessary but 
slows down the calculations by a factor of up to 100 !!!

Use:   init -prec 1 (for a reasonable accuracy) or 2 (for very good 
convergence, which I would use only at the very end once the surface is 
fully relaxed)

For a magnetic surface it is not unusually that the scf needs many 
cycles (depending on your case even 100 or so), and the minimization may 
take several hundreds of cycles (therefore one does it with low RKMAX 
and fewer k-points).

Am 28.03.2024 um 06:34 schrieb harriron at sjtu.edu.cn:
> Dear all,
> I am doing the optimization with the surface slab of KFe2As2.
> When i use "run_lapw",it can be convergent with fc 20 in 21 cycles.
> And when i use "runsp_lapw",it doesnt converge in fc 20.
> 
> I would like to know if spin polarization calculation is necessary for 
> optimization.
> They both do the initialization.
> "Init_lapw -b -vxc 13 -ecut -10 -rkmax 9.5 -numk 100 -fermit 0.002"
> 
> "Init_lapw -b -vxc 13 -ecut -10 -rkmax 9.5 -numk 100 -fermit 0.002 -sp"
> With regards.
> Harriron
> 
> 
> 发自我的手机
> 
> _______________________________________________
> 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

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