In which orbital is an excess electron placed in charged system?

[Fernan Saiz]

Hi Matt,
Thanks for your answers. 

Regarding to my second question, is it sensible using CP2k to correct the 
arbitrary energy levels obtained with the POISSON solver by running a 
second simulation with my system containing a large vacuum gap? I guess I 
could use the difference of the lowest core levels obtained in bulk and in 
the bulk-like region with the vacuum gap to correct these arbitrary energy 
levels.

Best regards,
 - Fernan

On Wednesday, October 11, 2017 at 8:52:20 PM UTC+1, Matt W wrote:
>
> Hi Fernan,
>
>> I would like to know how CP2k models an excess electron placed in a 
>> neutral system with N electrons by writing CHARGE -1.0 in the section 
>> CP2K_INPUT / FORCE_EVAL / DFT. Is this excess electron placed in the LUMO 
>> N-orbital?
>>
>> this isn't really how things work. The KS equations are solved, then 
> filled up obeying the Aufbau principle. If the excess charge doesn't 
> disturb the existing charge density, then it would be in an orbital very 
> similar to the previous LUMO.
>  
>
>> In addition, does cp2k correct the eigenvalues to the vacuum level? Or do 
>> they need to be corrected after the simulation is completed? Is there any 
>> reference?
>>
>
> this depends on you boundary conditions (POISSON solver). In a periodic 
> calculation there is no vacuum, or anything outside the system, and the 
> energy levels are arbitrary. In a surface or molecular system (without 
> dipole) the WAVELET solver will give the levels relative to zero.
>
> Matt 
>
>>
>> Best regards,
>>  - Fernan Saiz, PhD
>> Department of chemistry
>> Imperial College London
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20171012/9b526cfc/attachment.htm>