[Pw_forum] Slab optimization with -24kbar stress!!!
David Foster
davidfoster751 at yahoo.com
Thu Dec 4 19:26:16 CET 2014
Dear users
I have used 5.1.1 version to optimize conventional cell of CeO2 with 12 atoms.
I used PSLIBRARY to use USPP for Ce and O atoms. Here is my input for CeO2 conventional cell:
==================
&CONTROL
title = 'CeO2'
calculation = 'vc-relax'
restart_mode = 'from_scratch'
outdir = './ceo2_vc_relax'
pseudo_dir = './'
prefix = 'ceo2'
disk_io = 'default'
verbosity = 'default'
etot_conv_thr=1.0D-9
forc_conv_thr=1.0D-4
nstep=1000
wf_collect=.true.
/
&SYSTEM
ibrav = 0
nat = 12
ntyp = 2
ecutwfc = 173
ecutrho = 692
starting_magnetization(1)=0.5
starting_magnetization(3)=0.5
nspin=2
occupations='smearing'
degauss=0.001
smearing='mv'
nbnd=82
/
&ELECTRONS
electron_maxstep = 1000
conv_thr = 1.0D-10
mixing_mode = 'plain'
mixing_beta = 0.4
mixing_ndim = 8
diagonalization = 'david'
/
&IONS
ion_dynamics = 'bfgs'
/
&CELL
cell_dynamics = 'bfgs'
/
ATOMIC_SPECIES
O 15.9989995956 O.pbe-nl-rrkjus_psl.1.0.0.UPF
Ce 140.1199951172 Ce.pbe-spdfn-rrkjus_psl.1.0.0.UPF
ATOMIC_POSITIONS crystal
O 0.2500000000000000 0.2499999999999999 0.2499999999999999
O 0.7500000000000000 0.7499999999999999 0.2499999999999999
O 0.7500000000000000 0.2499999999999999 0.7499999999999999
O 0.2500000000000000 0.7499999999999999 0.7499999999999999
O 0.2500000000000000 0.2499999999999999 0.7499999999999999
O 0.7500000000000000 0.7499999999999999 0.7499999999999999
O 0.7500000000000000 0.2499999999999999 0.2499999999999999
O 0.2500000000000000 0.7499999999999999 0.2499999999999999
Ce 0.0000000000000000 0.0000000000000000 0.0000000000000000
Ce 0.0000000000000000 0.4999999999999999 0.4999999999999999
Ce 0.5000000000000000 0.0000000000000000 0.4999999999999999
Ce 0.5000000000000000 0.4999999999999999 0.0000000000000000
K_POINTS automatic
3 3 3 0 0 0
CELL_PARAMETERS
10.225308105044215 0.000000000000000 0.000000000000000
0.000000000000000 10.225308105044215 0.000000000000000
0.000000000000000 0.000000000000000 10.225308105044215
====================================================
Everything is OK for conventional cell vc-relax command (very low force and stress).
My problem is for slab optimization. I used following command to relax 2*2 110 surface slab with 6 layers and 4 bottom layers fixed.
You can see my input:
===================================================
&CONTROL
title = 'ceo2_110'
calculation = 'relax'
restart_mode = 'from_scratch'
outdir = './ceo2_110'
pseudo_dir = './'
prefix = 'ceo2_110'
disk_io = 'default'
verbosity = 'default'
tstress = .true.
tprnfor = .true.
etot_conv_thr=1.0D-9
forc_conv_thr=1.0D-4
nstep=1000
wf_collect=.true.
/
&SYSTEM
ibrav = 0
nat = 72
ntyp = 2
ecutwfc = 173
ecutrho = 692
starting_magnetization(1)=0.5
starting_magnetization(3)=0.5
nspin=2
occupations='smearing'
degauss=0.001
smearing='mv'
nbnd=490
/
&ELECTRONS
electron_maxstep = 1000
conv_thr = 1.0D-10
mixing_mode = 'plain'
mixing_beta = 0.4
mixing_ndim = 8
diagonalization = 'david'
/
&IONS
ion_dynamics = 'bfgs'
/
ATOMIC_SPECIES
O 15.9989995956 O.pbe-nl-rrkjus_psl.1.0.0.UPF
Ce 140.1199951172 Ce.pbe-spdfn-rrkjus_psl.1.0.0.UPF
ATOMIC_POSITIONS crystal
O 0.1250000000000000 -0.0000000000000000 0.0000000000000003 0 0 0
O 0.1250000000000000 -0.0000000000000000 0.1562436466930924 0 0 0
O 0.1250000000000001 0.0000000000000001 0.3124872933861844
O 0.3750000000000002 0.2499999999999999 0.2343654700396384 0 0 0
O 0.3750000000000002 0.2499999999999999 0.0781218233465463 0 0 0
O 0.3750000000000002 0.2500000000000000 0.3906091167327304
O 0.3750000000000002 0.0000000000000001 0.0000000000000003 0 0 0
O 0.3750000000000001 -0.0000000000000002 0.1562436466930924 0 0 0
O 0.3750000000000003 0.0000000000000001 0.3124872933861844
O 0.1250000000000001 0.2499999999999999 0.2343654700396383 0 0 0
O 0.1250000000000001 0.2500000000000001 0.0781218233465462 0 0 0
O 0.1250000000000001 0.2500000000000000 0.3906091167327304
O 0.6250000000000002 -0.0000000000000001 0.0000000000000003 0 0 0
O 0.6250000000000002 -0.0000000000000001 0.1562436466930924 0 0 0
O 0.6250000000000002 0.0000000000000000 0.3124872933861844
O 0.8750000000000002 0.2499999999999998 0.2343654700396384 0 0 0
O 0.8750000000000002 0.2499999999999998 0.0781218233465463 0 0 0
O 0.8750000000000002 0.2499999999999999 0.3906091167327304
O 0.8750000000000002 -0.0000000000000001 0.0000000000000003 0 0 0
O 0.8750000000000001 -0.0000000000000001 0.1562436466930924 0 0 0
O 0.8750000000000004 0.0000000000000000 0.3124872933861845
O 0.6250000000000002 0.2499999999999999 0.2343654700396384 0 0 0
O 0.6250000000000002 0.2500000000000002 0.0781218233465463 0 0 0
O 0.6250000000000002 0.2500000000000001 0.3906091167327304
O 0.1250000000000000 0.4999999999999999 0.0000000000000003 0 0 0
O 0.1250000000000000 0.5000000000000001 0.1562436466930923 0 0 0
O 0.1250000000000001 0.5000000000000001 0.3124872933861844
O 0.3750000000000002 0.7500000000000000 0.2343654700396384 0 0 0
O 0.3750000000000002 0.7499999999999999 0.0781218233465463 0 0 0
O 0.3750000000000002 0.7500000000000000 0.3906091167327304
O 0.3750000000000002 0.5000000000000001 0.0000000000000002 0 0 0
O 0.3750000000000001 0.4999999999999999 0.1562436466930924 0 0 0
O 0.3750000000000004 0.5000000000000001 0.3124872933861843
O 0.1250000000000001 0.7500000000000000 0.2343654700396383 0 0 0
O 0.1250000000000001 0.7500000000000001 0.0781218233465462 0 0 0
O 0.1250000000000001 0.7500000000000002 0.3906091167327304
O 0.6250000000000003 0.4999999999999999 0.0000000000000002 0 0 0
O 0.6250000000000001 0.5000000000000001 0.1562436466930923 0 0 0
O 0.6250000000000003 0.5000000000000001 0.3124872933861844
O 0.8750000000000002 0.7500000000000000 0.2343654700396384 0 0 0
O 0.8750000000000002 0.7499999999999999 0.0781218233465463 0 0 0
O 0.8750000000000002 0.7500000000000001 0.3906091167327304
O 0.8750000000000002 0.5000000000000001 0.0000000000000002 0 0 0
O 0.8750000000000000 0.4999999999999999 0.1562436466930924 0 0 0
O 0.8750000000000006 0.5000000000000001 0.3124872933861844
O 0.6250000000000002 0.7500000000000001 0.2343654700396384 0 0 0
O 0.6250000000000002 0.7500000000000001 0.0781218233465462 0 0 0
O 0.6250000000000002 0.7500000000000002 0.3906091167327305
Ce 0.0000000000000000 -0.0000000000000000 0.0781218233465463 0 0 0
Ce 0.0000000000000000 -0.0000000000000000 0.2343654700396383 0 0 0
Ce 0.0000000000000001 -0.0000000000000001 0.3906091167327305
Ce 0.2500000000000002 0.2500000000000000 0.0000000000000002 0 0 0
Ce 0.2500000000000002 0.2499999999999999 0.3124872933861844
Ce 0.2500000000000002 0.2500000000000001 0.1562436466930924 0 0 0
Ce 0.5000000000000001 0.0000000000000000 0.0781218233465463 0 0 0
Ce 0.5000000000000001 0.0000000000000000 0.2343654700396384 0 0 0
Ce 0.5000000000000001 -0.0000000000000001 0.3906091167327304
Ce 0.7500000000000003 0.2500000000000001 0.0000000000000002 0 0 0
Ce 0.7500000000000003 0.2499999999999999 0.3124872933861844
Ce 0.7500000000000003 0.2500000000000001 0.1562436466930924 0 0 0
Ce 0.0000000000000000 0.5000000000000001 0.0781218233465463 0 0 0
Ce 0.0000000000000000 0.5000000000000001 0.2343654700396383 0 0 0
Ce 0.0000000000000001 0.4999999999999999 0.3906091167327304
Ce 0.2500000000000001 0.7500000000000001 0.0000000000000001 0 0 0
Ce 0.2500000000000001 0.7500000000000000 0.3124872933861844
Ce 0.2500000000000001 0.7500000000000001 0.1562436466930923 0 0 0
Ce 0.5000000000000000 0.5000000000000000 0.0781218233465463 0 0 0
Ce 0.5000000000000002 0.5000000000000001 0.2343654700396384 0 0 0
Ce 0.5000000000000002 0.5000000000000000 0.3906091167327304
Ce 0.7500000000000002 0.7500000000000001 0.0000000000000001 0 0 0
Ce 0.7500000000000002 0.7500000000000000 0.3124872933861844
Ce 0.7500000000000002 0.7500000000000001 0.1562436466930923 0 0 0
K_POINTS automatic
1 1 1 1 1 1
CELL_PARAMETERS Bohr
20.556138517348749 0.000000000000002 -0.000000000000004
0.000000000000000 14.535384940627280 0.000000000000002
-0.000000000000000 -0.000000000000000 46.515123169068765
====================================================================
SCF in each DFGS iteration converges in about 200-300 steps.
My main problem is final section of the output:
======================
.
.
.
atom 71 type 2 force = 0.00000009 -0.00000005 -0.00000464
atom 72 type 2 force = -0.00000004 0.00000118 0.00769631
Total force = 0.000043 Total SCF correction = 0.000001
entering subroutine stress ...
negative rho (up, down): 7.323E-04 7.323E-04
total stress (Ry/bohr**3) (kbar) P= -24.51
-0.00007691 0.00000000 -0.00000000 -11.31 0.00 -0.00
0.00000000 -0.00032173 0.00000000 0.00 -47.33 0.00
-0.00000000 0.00000000 -0.00010117 -0.00 0.00 -14.88
bfgs converged in 19 scf cycles and 18 bfgs steps
(criteria: energy < 1.0E-09, force < 1.0E-04)
End of BFGS Geometry Optimization
.
.
==================================
As you see final stress is -24.51 kbar ??????!!!!!!!!!!!!!!
As you know, for slab, we don't used "vc-relax", and only optimize ions positions. What should I do to reduce the stress on the slab?
Regards
David Foster
Ph.D. Student of Chemistry
More information about the users
mailing list