[QE-users] not orthogonal operation in dos.x run only

[Paolo Giannozzi]

On Fri, Oct 16, 2020 at 9:39 PM Lucas Dória <lucasdoriacarvalho at hotmail.com>
wrote:

>
> So it is possible for a system to give satisfactory results in
> optimization calculations, and yet show symmetry problems in further
> routines?
>

The check that fails in your case is performed only in variable-cell MD and
when data is read for further processing. So, yes, when you read your data
you may discover that some symmetries that were assumed to be present
during the scf calculation are no longer found, according to the criteria
used to decide whether a symmetry is present or not. This typically happens
when  your system is at the border between symmetric and not symmetric.

Also, my job using bands.x with the optimized structure for WS2 finished
> without getting symmetry issues, which does not occur for dos.x and
> projwfc.x calculations.
>

I don't think this can happen, because ALL codes that read data from the
scf calculation perform the same check, so bands.x will fail as well.
Anyway: you can just remove the check by deleting the line that calls
"checkallsym" in PW/src/pw_restart.f90. It's a dumb check that makes more
harm than good.

Paolo


> I will work on your tip and try to see if I can solve this problem by
> re-submitting relax calculations with better parameters, but it is yet
> strange to me why the 'not orthogonal operation' crash doesn't appear in
> vc-relax calc, neither in scf and nscf calculations, and we just take note
> of it in dos.x and projwfc.x routines.
>
> With regards,
>
> Lucas.
>
> ------------------------------
> *De:* users <users-bounces at lists.quantum-espresso.org> em nome de Paolo
> Giannozzi <p.giannozzi at gmail.com>
> *Enviado:* sábado, 3 de outubro de 2020 13:06
> *Para:* Quantum ESPRESSO users Forum <users at lists.quantum-espresso.org>
> *Assunto:* Re: [QE-users] not orthogonal operation in dos.x run only
>
> Obscure symmetry problems almost invariably come from the usage of "free"
> Bravais lattice (ibrav=0) and almost-symmetric-but-not-enough lattice
> vectors, or by almost-symmetric-but-not-enough atomic positions
>
> Paolo
>
> On Sat, Oct 3, 2020 at 6:36 PM Lucas Dória <lucasdoriacarvalho at hotmail.com>
> wrote:
>
> Dear all,
>
> I ran into this problem while doing a dos.x calculation (nscf and scf
> calculations finished succesfully):
>
> *'from checkallsym : error # 2*
> *not orthogonal operation'*
>
> The structure is a WS2 bilayer. (WS2 is a layered material, we optimized
> bulk structure and then added vaccum in z-direction to get bilayer
> structure.)
>
> When i looked up for possible reasons and solutions, I found this problem
> only occuring in vc-relax calculations. Our group did in fact optimized the
> hexagonal structure, but vc-relax calculations finished succesfully.
>
> I add that we already studied electronic properties of another very
> similar hexagonal structure, of the same point group, namely MoS2, and we
> didn't get symmetry issues. (using v5.4 at the time)
>
> So, I don't know why this problem of 'not orthogonal operation' is
> happening to WS2, and just in dos.x calculation, if it's a crash due to
> symmetry breaking. We think that the only major differences between WS2 and
> MoS2 are that W is heavier than Mo, and the PP used. But we are using an
> optimized ecutwfc.
>
> Thanks in advance. QE-version used: 6.5.
>
> Lucas.
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
>
>
>
> --
> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
>
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users



-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20201017/e9fe6d29/attachment.html>