[Wien] Problem with LDA+U with Core Hole

[Peter Blaha]

It depends on the case if you should add the electron in case.inm or 
case.in2.
When the excitation in the experiment really goes mainly into states 
just above EF, I'd put this electron into in2; if the electron goes 
somewhere else, I'd use inm.
Eg:  NiO:  O-K edge goes into unoccupied O-2p states, but NiO has very 
little O-2p character in the conduction band. So when you add the e in 
case.in2, you will occupy mainly Ni-d, which is not what happens in 
experiment. So put it into case.inm. On the other hand, for a Ni-L edge, 
I'd put it into the case.in2 file.

In any case, the effect is usually small and supercell size might be 
more important.

On 11/5/18 2:07 PM, Laurence Marks wrote:
>  From what you wrote it appears that you are adding the core hole at the 
> start of the calculation. I suggest that instead you first converge your 
> supercell as a normal calculation, save it, then add your core hole (or 
> fractional core hole, e.g. 1/2 for Slater method).
> 
> N.B., the background charge is added to case.inm, to my knowledge it is 
> incorrect to change case.in2(c).
> 
> N.N.B., remember that for a spin-polarized calculation you need to do 
> two calculations, one with a core hole in case.incup and one with a core 
> hold in case.incdn. You will also need to be careful with telnes since 
> it does not seem to automatically pickup the case.inc(up/dn) so you have 
> to do it by hand.
> 
> On Mon, Nov 5, 2018 at 5:17 AM catalina coll <ccoll at el.ub.edu 
> <mailto:ccoll at el.ub.edu>> wrote:
> 
>     Dear all,
> 
>     I'm trying to perform a LDA+U with Core Hole calculation of
>     rombohedral LuFe2O4 using WIEN2k_13.1.
> 
>     First, I did the LDA+U calculation without core hole and it converged.
> 
>     Then, I wanted to improve the results for the ELNES with core hole
>     and I've followed the steps of the UG, but the simulations stops
>     because of the presence of ghost bands.
> 
>     Basically, I followed this steps:
>     1. make a supercell and remove symmetry by the change of the name of
>     one atom.
>     2. init_lapw
>     3. add an electron on .in2c (or in .inm) and remove one on .inc
>     4. runsp_c_lapw -orb -p...
> 
>     In order to solve the problem I have been trying to modifying the
>     .in1c like is said on the FAQ and also I have tried changing the
>     size of my supercell (for a smaller supercell the ghost band appear
>     later).
> 
>     Do you have any suggestions for me?
> 
>     Thanks a lot,
> 
>     -- 
> 
> 
>     Catalina Coll
> 
>     PhD Candidate
> 
>     LENS - Laboratory of Electron Nanoscopy MIND - Micro-Nanotechnology
>     and Nanoscopies for electrophotonic Devices IN2UB - Institute of
>     Nanoscience and Nanotechnology
> 
>     Departament d'Enginyeria Electrònica i Biomèdica - Universitat de
>     Barcelona c/ Martí i Franquès 1 08028 Barcelona Tel: (+34) 93 403 91 75
> 
> 
> 
> -- 
> Professor Laurence Marks
> "Research is to see what everybody else has seen, and to think what 
> nobody else has thought", Albert Szent-Gyorgi
> www.numis.northwestern.edu <http://www.numis.northwestern.edu> ; 
> Corrosion in 4D: MURI4D.numis.northwestern.edu 
> <http://MURI4D.numis.northwestern.edu>
> Partner of the CFW 100% program for gender equity, 
> www.cfw.org/100-percent <http://www.cfw.org/100-percent>
> Co-Editor, Acta Cryst A
> 
> _______________________________________________
> 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
--------------------------------------------------------------------------