[Wien] Slab convergence

[Peter Blaha]

40 iterations is not much for a problem like this. I have no idea why 
your calculation went strange at the end of the first run, and I doubt 
that anyone can guess.


Exactly. It looks quite reasonable until the big jumps in :ENE at the 
end. I'd either continue, or restart from scratch. You must have done 
something strange.....

>
> Third, why are you using the bulk lattice parameters? Particularly for 
> surfaces you need to use the DFT ones, especially in the surface plane.
>
Well, we do not know what the motivation and purpose of this calculation 
is. But yes, for a surface one would usually use the theoretical a,b 
lattice parameters, not the experimental ones (and for a TMDC you could 
keep c constant, since PBE will not do a good job).
> Fourth, why are you not relaxing? Not only is this right, big 
> movements will tell you that your slab is wrong. Note: you cannot 
> relax with -so.
TMDCs have van der Waals bonds between layers. So a relaxation makes 
only sense with a vdW functional. PBE  may drive the layers apart.


PS: A surface should always be done with TEMP. Actually, init_lapw 
should automatically switch to TEMP if it detects a surface. You do not 
have "tetrahedra" but only degenerate triangles with a surface slab.


> --
> Professor Laurence Marks
> Department of Materials Science and Engineering, Northwestern University
> www.numis.northwestern.edu <http://www.numis.northwestern.edu>
> "Research is to see what everybody else has seen, and to think what 
> nobody else has thought" Albert Szent-Györgyi
>
> On Mon, Feb 20, 2023, 05:09 pluto via Wien 
> <wien at zeus.theochem.tuwien.ac.at> wrote:
>
>     Dear Prof. Marks, dear All,
>
>     I am calculating a large slab of one of the 1T TMDC materials, 30
>     unique
>     atoms (this makes it 20 tri-layers, i.e. 20 vdW layers).
>
>     The slab of the same material of half size, 15 unique atoms (10
>     tri-layers) converged well.
>
>     I am using TEMP 0.002 (first I tried with TETRA, but restated with
>     TEMP
>     after some iterations). It is a spin-polarized calculation with SOC.
>     Inversion is present. Experimental lattice constant, slab made
>     with "x
>     supercell", no relaxation. Slab looks good in xcrysden ("x supercell"
>     sometimes puts surface atoms wrongly, but I corrected it.). I used 30
>     Bohr vacuum between layers. I use 36 k-points (klist pasted below).
>     Other parameters are WIEN2k default.
>
>
>
>
> _______________________________________________
> 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: +43-158801165300
Email:peter.blaha at tuwien.ac.at           
WWW:http://www.imc.tuwien.ac.at       WIEN2k:http://www.wien2k.at
-------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20230220/4ddeb881/attachment.htm>