[DFTB-Plus-User] Am I fully optimizing the charges of my structure?

[Bálint Aradi]

Dear Alathea,

> First set of questions: If I converge this geometry at a Fermi
> Temperature of 100 K, is that at a true energetic minimum? Will giving a
> Fermi Temperature above 0 K generate an inaccurate band structure or
> density of states, if the charges aren’t fully optimized?  

If a structure, especially a periodic one, has partially filled bands,
electronic temperature is very often the only way to get it converged.
What you get then, is the minimum of the free energy E-TS. If the
temperature is moderate, it is usually very-close to the minimum of the
energy.

As for the error you get with the band structure: There was a type in
the MaxAngularMomentum in your band structure calculation. If you fix
that (O has basis "p", not "s"), the nr. of electrons is correct.

> Another issue that I believe I’m having is that, despite the SCC and
> geometry converging previously and reading those output files into the
> band structure script, in the new script, which is reading in charges
> from the converged calculation, the SCC doesn’t converge in one
> iteration for the band structure as I expected it to, according to
> the DFTB+ Band Structure
> <https://dftbplus-recipes.readthedocs.io/en/latest/basics/bandstruct.html> recipe.
> In the example file attached, when I added the “Charge” command, it took
> 35 iterations for the SCC to converge (Band1.log).  

Note, that when doing band structure calculation, the MaxSccIterations
should be set to 1. (You want to diagonalize the Hamiltonian only once,
with the charges read in.)

  Best regards,

  Bálint

-- 
Dr. Bálint Aradi
Bremen Center for Computational Materials Science, University of Bremen
http://www.bccms.uni-bremen.de/cms/people/b-aradi/


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20220221/5d7d8d3c/attachment.sig>