[CP2K-user] Problem with K-edge X-Ray absorption calculation

[Marcella Iannuzzi]

Hi Patrick

My guess is that the problem is due to the localisation performed on all MOS
On the other hand there is no not need to localise all the states. I would 
restrict it to the core ones, so that the algorithm can easily identify the 
1s of the O you indicate.
This can be done by activating the keyword STATE_SEARCH to restrict the 
localisation only to a given number of states, starting from the lowest in 
energy. 
In this case it should be sufficient to give a number of states  equal to 
the number of atoms, since the 1s of Si and O should be the lowest in 
energy.

Kind regards
Marcella

On Monday, May 3, 2021 at 10:17:25 AM UTC+2 Patrick wrote:

> Hello everyone,
>
> I tried doing K-edge X-Ray absorption calculations with cp2k 5.1 (since I 
> started the project with this version, but version 7 gives me the same 
> error) of a Si/SiO2 system surrounded by water. The structure is chosen 
> from an MD run. For the Silicon atoms everything works fine, but when I try 
> to do a calculation on the oxygen atoms of the water molecules most of them 
> do not work (a few do) and crash with the following error message:
>
>
>  *******************************************************************************
>  *   
> ___                                                                       *
>  *  /   
> \                                                                      *
>  * 
> [ABORT]                                                                     
> *
>  *  \___/    A wrong state has been selected for excitation, check the 
> Wannier *
>  *    |                                   
> centers                              *
>  *  
> O/|                                                                        *
>  * /| 
> |                                                                        *
>  * / \                                                      
> xas_methods.F:1590 *
>
>  *******************************************************************************
>
> I attached the structure, one input and output as well. It is the first 
> system I encounter this problem on and any help would be greatly 
> appreciated. 
>
> Best regards
> Patrick
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210503/0f7f8303/attachment.htm>