[CP2K-user] MOs in Molden Format Fail the Normalization Check

[Brandon Meza González]

Dear, Arshad

I am facing the same problem, did you find a way to fix it?

Best Regards,

Brandon

El viernes, 19 de julio de 2019, 4:57:17 (UTC-5), Arshad Mehmood escribió:
>
> Dear Developers,
> I need molecular orbitals in MOLDEN format for post-processing (not just 
> visualization)  to calculate Orbital Overlap Distance Function 
> <https://onlinelibrary.wiley.com/doi/abs/10.1002/anie.201702715>. 
> I have modified the lines in "molden_utlis" as recommended in the thread "MOS 
> in Molden Format 
> <https://groups.google.com/forum/#!msg/cp2k/CYGzR4T2s5U/6yS0HXWxBAAJ>". 
> Namely;
>
> WRITE (iw, fmtstr1) irow_in+orbital - 1 , mo_coeff(orbmap(orbital))
>
> I am getting .MOLDEN output file containing orbitals in correct format.
> However, the molecular orbitals fail the normalization check. The sum of 
> MOs occupancies are largely different from the analytically integrated 
> number of electrons. For small molecules e.g. water at  PADE 
> DZVP-GTH-PBE/GTH-PBE the difference is 0.0021, but for system like 6x6x1 
> graphene super-cell, the difference reaches to 10.348.
> I shall be very much thankful if you suggest me the reason and the 
> possible fix. I need the correct number of electrons for my calculations.
>
> Best Regards,
>
> Arshad Mehmood
> Texas Christian University.    
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200528/7fcbf2b7/attachment.htm>