[CP2K-user] DOS, GAPW, Iron ions using EMSL_BASIS_SET

[Marcella Iannuzzi]

Dear Ivan

The basis sets that you mentioned are typically generated for molecular 
systems and are not optimal for condensed matter. 
In general less diffuse BS are more stable. Search for BS that are 
optimised for DFT and combined with auxiliary BS if you use hybrid 
functionals. 
https://pubs.acs.org/doi/pdf/10.1021/acs.jpca.7b04760
http://www.psicode.org/psi4manual/master/basissets_byfamily.html

Regards
Marcella




On Sunday, September 13, 2020 at 10:53:33 PM UTC+2 ig... at gmail.com wrote:

> Dear CP2K users
>
> I would like to calculate the DOS for Iron ions in aqueous solutions of 
> different salts. Since I am interested in both core and band levels, I am 
> planning to do an all-electron (energy) calculation at PBE0-D3 level from 
> different snapshots taken from an MD trajectory using GAPW approach 
> exploiting EMSL_BASIS_SET
>
> I gave a look at the EMSL_BASIS_SET file and I saw many possible choices 
> for Iron. I counted 11 of them. Ahlrichs-def2-QZVP, 6-31G*, 6-31G**, 
> Ahlrichs-VDZ, Ahlrichs-VTZ, Ahlrichs-pVDZ, Ahlrichs-TZV, Ahlrichs-pTZV, 
> Roos-ADZ-ANO, Roos-ATZ-ANO, NASA-Ames-cc-pVTZ
>
> I tried a couple of them: for example, with 6-31G** I was able to converge 
> the SCF cycle, with Ahlrichs-def2-QZVP no.
>
> I would like to ask you if you have any advice based on your experience 
> for a reasonable basis set choice to start (the one above or even others). 
> I look in the literature: I am familiar with some of the basis set listed 
> above, others are new for me.
>
> Thank you very much for any possible advice
>
> Best
>
> Ivan
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200914/db1bdab3/attachment.htm>