[CP2K-user] contribution of atomic orbitals to molecular orbitals

[Vladimir Rybkin]

Dear Fernan,

the MO coefficients - that's what you ask for - are huge matrices. Looking 
at them will barely give you an idea of the degree of localisation. 

So, if you want local orbitals, you can do MAO, or else, you can do Wannier 
localization. Electron localisation function can be an option as well. It 
depends on what exactly you want. 

Yours,

Vladimir



понедельник, 1 июня 2020 г., 10:54:26 UTC+2 пользователь Fernan Saiz 
написал:
>
> Dear all,
> I would like to know if CP2k can print the contribution of every atomic 
> orbital to each molecular orbital in a given system. Plotting molecular 
> orbitals in cube files can give a qualitatively idea if electrons are 
> localised, but it can be inaccurate if these particles are delocalised in 
> condensed systems. From the manual, it might look like MAO might be able to 
> do this job.
>
>
> Best regards,
>  - Fernan Saiz, PhD
>  Marie Skłodowska-Curie individual fellow
>  Institut de Ciència de Materials de Barcelona (ICMAB)
>  Spanish National Research Council (CSIC)
>  Campus de la UAB
>  Bellaterra, 08193, Spain    
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200602/12ca7035/attachment.htm>