[Wien] Plane-wave cutoff in W2K

[Marcos Veríssimo Alves]

Dear Prof. Blaha,

Thanks a lot for the answer - from someone originally from the
pseudopotential community :-) . I suspected that it was RMT(min) that would
define the cutoff, but I had to be sure.

Best regards,

Marcos

On Mon, Jun 6, 2011 at 2:27 PM, Peter Blaha <pblaha at theochem.tuwien.ac.at>wrote:

> The energy cutoff for the planewave representation in the interstitial is
> $E_{max}^{wf}=19$ Ry for the wave functions and $E_{max}^{pot}=196$ Ry for
> the potential."
>
> A specification like this comes from the Pseudopotential community. We use
> different values:
>
>
> The cutoff for the potential (and charge density) is specified in
> case.in2(c). By default we use GMAX=12, while in this paper
> GMAX=14 (E=k^2; 14^2=196 Ry) was used.
>
> The cutoff for the wave functions is specified in case.in1(c), but as
> RMT(min) * Kmax
> So the knowledge of "19 Ry" cutoff alone is not sufficient. You also need
> to know what their smallest RMT was.
> Then RKmax = RMT(min) * sqrt(19).   Eg. an RMT=1.6 (for Oxygens) gives our
> default RKMax=7
>
>
>  The authors specify their convergence parameters in the following way:
>>
>> "Inside the muffin tins wave functions are expanded in spherical harmonics
>> up to $l_{max}^{wf}=10$ and nonspherical contributions to the electron
>> density and potential up to
>> $l_{max}^{pot}=6$ are used. The energy cutoff for the planewave
>> representation in the interstitial is $E_{max}^{wf}=19$ Ry for the wave
>> functions and $E_{max}^{pot}=196$ Ry for
>> the potential."
>>
>> I would like to specify my parameters in the same way. For $l_{max}^{wf}$
>> and $l_{max}^{pot}$, these parameters are found in case.in <
>> http://case.in>:
>>
>>
>> 8.00       10    4 (R-MT*K-MAX; MAX L IN WF, V-NMT)
>>
>> and the last two would give me $l_{max}^{wf}$ and $l_{max}^{pot}$.
>> However, I am in doubt on how to find the PW cutoff for the wavefunctions in
>> the interstitials and for the
>> potential (or how to determine these cutoffs, if they can't be readily
>> found in the output files).
>>
>> For the wavefunctions, it is somewhat straightforward: kmax^2 gives the
>> cutoff energy in Ry. I suppose I could just take the smallest cutoff (since
>> there will be different kmax
>> for the different elements) as a specification, or is the maximum/minimum
>> value of kmax used for all elements, instead?
>>
>> For the expansion of potentials in plane waves, I'd suppose the energy
>> cutoff would be simply gmax^2 (as specified in case.in2), but I'd like to
>> confirm it, even under the risk of
>> having made a stupid question :)
>>
>> Any help will be greatky appreciated.
>>
>> Thanks in advance,
>>
>> Marcos
>>
>> ---------------
>> Marcos Veríssimo Alves
>> Universidad de Cantabria, Spain
>>
>>
>>
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>
> --
>
>                                      P.Blaha
> --------------------------------------------------------------------------
> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
> Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
> Email: blaha at theochem.tuwien.ac.at    WWW:
> http://info.tuwien.ac.at/theochem/
> --------------------------------------------------------------------------
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20110606/16f82220/attachment.htm>