[CP2K-user] [CP2K:20149] Re: XAS Calculation

[Marcella Iannuzzi]

Hello Gerard, 

There are several problems in the simulation you are running.
First of all, for Fe an all electron potential is indicated, but a basis 
set for the valence electron only is used. This is not going to work.
The ground state electronic structure calculation is, not surprisingly, 
converging very badly.
Not knowing the properties of your system, it is difficult to judge, what 
the optimal settings would be, in terms of PP, basis set, cutoff, spin 
polarization, etc
I suggest that for the PP description you use POTENTIAL_UZH and 
BASIS_MOLOPT_UZH libraries. For the ALL electron description check you can 
take from the EMSL_BASIS_SET or check in the literature what is suggested 
for K-edge spectra of transition metals. 
It is possible that also for the ground state one should use 
diagonalization and smearing to properly converge the SCF. 
For the TP method to obtain XAS it is not possible to use OT because the 
occupation number of the excited core state is modified, meaning that the 
diagonalization approach has to be used. Sometime G-space mixing and 
smearing help the convergence.
Consider also the possibility to compute XAS with XAS_TDP.

Regards
Marcella



On Monday, April 29, 2024 at 2:46:04 PM UTC+2 gpar... at gmail.com wrote:

> Hello everyone,
>
> I am trying to run a XAS calculation, however the convergence for the XAS 
> section is very bad and there is no way the SCF is converging. I have 
> reduced a lot the threshold just for testing but still not working. The 
> system I am using is a bit tricky, 13Fe in a cluster on a SiO2 surface. I 
> was wondering, is there any way to apply the CG minimizer for the SCF 
> section on XAS? If not, how we can improve the convergence for that section?
>
> Please find attached here the input and output for the calculation.
>
> Any suggestion for improving the calc is more than welcome!
> Thank you very much in advance.
>
> Gerard
>
>
>

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/8869f7b1-aadb-4f57-bed6-9df38dd60c4bn%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240430/f250741a/attachment.htm>