[Pw_forum] Convergence issue for C/BN slab
Biswajit Santra
bishalya at gmail.com
Mon Feb 20 20:47:14 CET 2017
Hello,
A few issues I could spot by looking at the input. Should use only 1
k point along the direction of the vacuum. Also, it will be more reasonable
if the number of k points in the direction of celldm(2) is about twice of
that in the direction of celldm(1). The ecutwfc is too large, 60-80 Ry
should be fine.
Best wishes,
Biswajit
Biswajit Santra
Mobile: +1-609-227-9202
http://www.princeton.edu/~bsantra/
On Mon, Feb 20, 2017 at 12:13 PM, Namita Narendra <nnarend at ncsu.edu> wrote:
> Hi,
>
> I am trying to relax a (111) diamond/cubic-BN slab and the total force on
> the atom is not converging. Initially, the scf cycle itself was not
> converging and I needed to introduce smearing with degauss=0.01 to achieve
> scf convergence. Now, the relaxation is not converging and the total force
> on the atom is oscillating. I have tried mixing of 'local-TF', reduced
> mixing_beta, used damped ion dynamics and tried LDA pseudopotential. None
> of this has helped with the convergence issue. Attached below is the input
> file used.
>
> &CONTROL
> calculation = 'relax' ,
> restart_mode = 'from_scratch' ,
> wf_collect = .true. ,
> outdir = '/share2/nnarend/espresso/c_bn1' ,
> pseudo_dir = '/home/nnarend/pseudo' ,
> prefix = 'C_BN_1' ,
> tstress = .true. ,
> tprnfor = .true. ,
> forc_conv_thr=1.0D-3,
> etot_conv_thr=1.0D-4
> /
> &SYSTEM
> ibrav = 8,
> celldm(1) = 8.291523,
> celldm(2)=0.57735,
> celldm(3)=6,
> nat = 24,
> ntyp = 3,
> ecutwfc=180.0,
> occupations='smearing',
> smearing='cold',
> degauss=0.01
>
> /
> &ELECTRONS
> conv_thr = 1.0d-8,
> mixing_mode = 'plain' ,
> mixing_beta = 0.3 ,
> diagonalization = 'david' ,
> /
> &IONS
> /
> ATOMIC_SPECIES
> B 10.811 B.pbe-nc.UPF
> N 14.007 N.pbe-nc.UPF
> C 12.011 C.pbe-nc.UPF
>
> ATOMIC_POSITIONS alat
> C 0.000000 0.000000 0.000000
> C 0.500000 0.288675 0.000000
> C 0.166667 0.288675 0.117851
> C 0.666667 0.000000 0.117851
> C 0.166667 0.288675 0.471404
> C 0.666667 0.000000 0.471404
> C 0.333333 0.000000 0.589255
> C 0.833333 0.288675 0.589255
> C 0.333333 0.000000 0.942808
> C 0.833333 0.288675 0.942808
> C 0.000000 0.000000 1.060659
> C 0.500000 0.288675 1.060659
> B 0.000000 0.000000 1.414212
> B 0.500000 0.288675 1.414212
> N 0.166667 0.288675 1.532063
> N 0.666667 0.000000 1.532063
> B 0.166667 0.288675 1.885616
> B 0.666667 0.000000 1.885616
> N 0.333333 0.000000 2.003467
> N 0.833333 0.288675 2.003467
> B 0.333333 0.000000 2.357020
> B 0.833333 0.288675 2.357020
> N 0.000000 0.000000 2.474871
> N 0.500000 0.288675 2.474871
>
> K_POINTS automatic
> 8 8 6 0 0 0
>
> Below is the force at the end of each bfgs cycle
>
> Total force = 0.492043 Total SCF correction = 0.000193
> Total force = 0.170844 Total SCF correction = 0.000253
> Total force = 0.049967 Total SCF correction = 0.000085
> Total force = 0.038773 Total SCF correction = 0.000075
> Total force = 0.042155 Total SCF correction = 0.000056
> Total force = 0.046807 Total SCF correction = 0.000072
> Total force = 0.044401 Total SCF correction = 0.000064
> Total force = 0.034652 Total SCF correction = 0.000095
> Total force = 0.033861 Total SCF correction = 0.000065
> Total force = 0.107334 Total SCF correction = 0.000129
> Total force = 0.071144 Total SCF correction = 0.000057
> Total force = 0.053109 Total SCF correction = 0.000077
> Total force = 0.058897 Total SCF correction = 0.000087
> Total force = 0.075144 Total SCF correction = 0.000061
> Total force = 0.109071 Total SCF correction = 0.000025
> Total force = 0.121561 Total SCF correction = 0.000042
> Total force = 0.128205 Total SCF correction = 0.000048
> Total force = 0.138713 Total SCF correction = 0.000070
>
> Thanks,
> Namita
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170220/262abc6e/attachment.html>
More information about the users
mailing list