[QE-users] Interested in difficult to converge systems

[Tamas Karpati]

Mam,
It depends much on what you want to calculate (geometry? energies?).
Try a few, start with PBE (with small model systems, yet including
possibilities for H bonds to form).
Test a few with nonlocal terms, eg. vdW-DF* and rVV10.
  t

On Thu, Jul 30, 2020 at 7:44 PM Matthew Marcus <mamarcus at lbl.gov> wrote:
>
> I've had similar issues with convergence on, for instance a Ni3F2(OH)4
> structure related to layered Ni(OH)2.  What functional would you suggest
> for that?
>         mam
>
> On 7/30/2020 10:44 AM, Tamas Karpati wrote:
> > Dear John,
> >
> > Currently a known bug of gradients from hybrid functionals provides
> > what you need. Try eg. B3LYP.
> > (If you are aiming at it from a theoretical point of view, i would
> > suggest weakly bonded, eg. vdW systems
> > or -to the contrary- strongly bonded systems where bond length changes
> > the type of bonding. The former is
> > a low gradient problem, the latter is coming from a perturbed chemical bonding.)
> >
> > HTH,
> >    t
> >
> > On Thu, Jul 30, 2020 at 7:13 PM John McFarland <jm18ec at my.fsu.edu> wrote:
> >>
> >> Dear all,
> >>
> >> I'm interested in systems where SCF has trouble converging to self consistency and instead keeps fluctuating. In particular I would really like any input files of such systems.
> >>
> >> Best Regards,
> >> John McFarland
> >> _______________________________________________
> >> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> >> users mailing list users at lists.quantum-espresso.org
> >> https://lists.quantum-espresso.org/mailman/listinfo/users
> > _______________________________________________
> > Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> > users mailing list users at lists.quantum-espresso.org
> > https://lists.quantum-espresso.org/mailman/listinfo/users
> >
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users