[Pw_forum] problem regrading convergence

Layla Martin-Samos lmartinsamos at gmail.com
Sun Apr 15 16:36:18 CEST 2012


Dear Bramha, infortunately this kin of behavior could happen, in particular
if the elctronic groundstate could have  two solutions. You can try
wdeducing to 0.1 mixing_beta and choosing mixing_mode="local-TF".

I hope it helps

bests

Layla

2012/4/15 bramha pandey <pandey.bramha at gmail.com>

> Dear all PW users,
> I m facing convergence problem with my LiInTe2 system. I have reduced the
> scf threshold, mixing from 0.7 to 0.8, ecut from 110Ry to 150Ry
> But i could not find my convergence after 100 iterations it was stop with
> message, convergence not achieved as my system it should not take much
> iteration might be i am  wrong. Also i was using Gamma point for fast
> convergence with CG diagonalization mehod
> Please look my input file....
> &control
>     calculation = 'scf'
>     restart_mode='from_scratch',
>     pseudo_dir = '  ',
>     outdir=' '
>     prefix = 'lit' ,
>     tprnfor = .true. ,
>  /
> &SYSTEM
>                        ibrav = 4,
>                    celldm(1) = 12.13,
>                    celldm(3) = 2.0,
>                          nat = 4,
>                         ntyp = 3,
>                      ecutwfc = 150,
>                      ecutrho = 550 ,
>  /
>  &ELECTRONS
>                     conv_thr = 1.0d-9 ,
>                  mixing_beta = 0.8 ,
>         mixing_ndim = 16,
>         diagonalization = 'cg',
>  /
>  &IONS
>                 ion_dynamics = 'bfgs' ,
>
>  /
>  &CELL
>                cell_dynamics = 'bfgs' ,
>                        press = 0 ,
>                        wmass = 40 ,
>             cell_factor = 2.0D0,
>  /
> ATOMIC_SPECIES
>    Li    6.94100  Li.pz-n-vbc.UPF
>    In  114.82000  In.pz-bhs.UPF
>    Te  127.60000  Te.pz-bhs.UPF
>
> ATOMIC_POSITIONS crystal
>    Li      0.000000000    0.000000000    0.000000000
>    In      0.50000000    0.50000000      0.50000000
>    Te      0.500000000    0.000000000    0.75
>    Te      1.0000000      0.50000000     0.25
> K_POINTS gamma
>
> Any valuable comment will be highly appreciated.
>
> --
> Thanks and Regards
> Bramha Prasad Pandey
> GLA University, Mathura.
> INDIA.
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20120415/f66ecc49/attachment.html>


More information about the users mailing list