[CP2K-user] [CP2K:22224] Re: Extrapolation formula for basis set convergence

[Frederick Stein]

Dear Riccardo,
Both is possible. Keep in mind that it is an approximation anyways. The 
correlation energies (without EXX) is described well by the given formula. 
The other contributions (SCF, EXX) usually decay exponentially but for 
simplicity, you may use the same formula as with RPA. As you can see, the 
difference is quite small. Be aware that there is a variety of other 
sources of errors (beyond-RPA corrections, vibrational corrections, 
geometry, experimental errors, core corrections). From my experience, 
higher order basis set corrections (X^-4-term) are roughly an order of 
magnitude lower than the basis set correction itself, so are also at the 
order of the deviation you observe between both results.
Best,
Frederick

Riccardo Pezzetta schrieb am Mittwoch, 6. Mai 2026 um 12:42:16 UTC+2:

> Sorry I wrote the wrong link for the article, the DOI is 10.1021/ct4002202
>
> Riccardo Pezzetta
>
> Il giorno mercoledì 6 maggio 2026 alle 12:37:59 UTC+2 Riccardo Pezzetta ha 
> scritto:
>
>> Dear all,
>> I am doing a calculation for the binding energy of the argon dimer with 
>> RPA and I used as primary basis sets aug-cc-pVTZ and aug-cc-pVQZ. In the 
>> CP2K maual I saw that I have to use the extrapolation formula E(X)=A+B/X^3 
>> for correlation energies, but I don't know how to extrapolate the EXX 
>> energy. In the article https://pubs.acs.org/journal/jctcce I saw that 
>> they use the extrapolation formula for the cohesive energies instead of 
>> correlation energies. So I tried to make the extrapolatoin in two ways:
>> 1) I calculated the binding energy of the dimer with TZ and QZ basis sets 
>> separately and made the cubic extrapolation with the resulting bonding 
>> energies
>> 2) I made the extrapolation with the cubic formula with absolute RPA and 
>> EXX energies for the dimer and the atom separately and after that I 
>> calculated the bonding energy with the extrapolated results.
>> With the first method I get E bonding = -8.4 meV, with the second one I 
>> get E bonding = -8.6 meV.
>> Are these procedures correct? Since the results are similar can I deduce 
>> that both methods are correct? In both cases the atom calculation has been 
>> made with a ghost atom with the same geometry of the dimer in order to take 
>> into account the basis set superposition error.
>> Kind regards;
>>
>> Riccardo Pezzetta
>
>

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/63b3b2a8-2995-4133-ac52-2c831e07e120n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20260506/8b9d52c3/attachment-0001.htm>