[QE-users] NEB calculation no symmetry found

[Stefano Baroni]

> On 13 Feb 2019, at 09:50, Offermans Willem <willem.offermans at vito.be> wrote:
> 
> Dear Quantum Espresso friends,

Dear Wilem:

> I made a mistake in my previous e-mail.
> The atom coordinates are not in Angstrom but in Bohr.
> 
> I have corrected this in my calculations, but still no symmetry found.

I would be surprised if you found the contrary. I cannot imagine that the point symmetry of any system could depend on units: do you?

> $ grep "ymmetr" pd_?/PW.out 
> pd_1/PW.out:     No symmetry found
> pd_2/PW.out:     No symmetry found
> pd_3/PW.out:     No symmetry found
> pd_4/PW.out:     No symmetry found
> pd_5/PW.out:     No symmetry found
> pd_6/PW.out:     No symmetry found
> pd_7/PW.out:     No symmetry found
> 
> So the question remains. What is the reason that the symmetry cannot be detected?
> 
> As an extra check I have made a scf calculation on the last image.
> 
> $ grep "Sym. Ops." pd.out 
>       2 Sym. Ops., with inversion, found ( 1 have fractional translation)
> 
> So pw.x can find the symmetry whereas neb.x cannot!

Just guessing: maybe for some reasons neb restricts its search to pure rotations (no fractional translations involved). Also, I have the impression that inversion symmetry does not help save any computer time, because in practice it has the same effects as time-reversal symmetry, which always holds.

Just my worthless 5 cents …
SB


— 
Stefano Baroni - Trieste —  http://stefano.baroni.me <http://stefano.baroni.me/> 




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20190213/c3d64bbf/attachment.html>