[QE-users] scf of organic molecule does not converge

[Stefano Baroni]

Smearing may alleviate convergence problems in any case where the latter are due to level crossing during the SCF cycle. Absolute energy errors are not necessarily a good metric of the accuracy of a pseudopotential calculation, as they depend sensitively on the details of the (pseudo) wave-function in the atomic core, which hardly affect the measurable chemico-physical properties of your system (such as e.g. forces and energy differences in general). SB

___
Stefano Baroni, Trieste -- http://stefano.baroni.me

> On 23 Jun 2021, at 17:26, Bidault, Xavier <xavbdlt at uic.edu> wrote:
> 
> 
> I would say that an energy cutoff of 25 Ry is way too small. I also work on organic molecules, and I use 90 Ry (and 900 Ry for ecutrho) with PSlib 1.0.0 US to get errors lower than 1 meV/atom.
> 
> @Paolo: I wonder how relevant it is to use smearing in an organic compound (unless it has conducting or semi-conducting capabilities). Could you tell why it would be something to do?
> 
> Xavier B.
> From: users <users-bounces at lists.quantum-espresso.org> on behalf of Paolo Giannozzi <p.giannozzi at gmail.com>
> Sent: Wednesday, June 23, 2021 2:04 AM
> To: Quantum ESPRESSO users Forum <users at lists.quantum-espresso.org>
> Subject: Re: [QE-users] scf of organic molecule does not converge
>  
> It works decently well if you add a smearing (and a few more bands, but this is done automatically):
>    occupations      = 'smearing'
>    smearing         = 'gaussian'
>    degauss          = 0.01
> 
> Paolo
> 
> On Wed, Jun 23, 2021 at 8:12 AM Haoran Wang <whr1130 at gmail.com> wrote:
> Hi QE users,
> 
> I have been trying to run the scf of a molecule below for many days. It never converged. The configuration is from Gaussian optimization, so it should be sensible. I did try other configurations obtained from MD simulations and still no convergence. I tried different pseudo-potentials (PBE-PAW, PBE-US, LDA-US) and also tried tuning the parameters (mixing_beta, mixing_mode, ecutrho...), still no convergence. Could somebody help me and see why this happens?
> 
> Thanks,
> Haoran Wang
> Utah State University
> 
> **************************************************************************************
> _______________________________________________
> 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 206, 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
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20210623/a8d38354/attachment.html>