[QE-users] Energy not decreasing when running vc-relax calculation with 'r2scan' functional
陈科帆
victor-chenkefan at 163.com
Thu Jun 20 08:50:07 CEST 2024
Hi there,
Thank you for your previous help on ld1.x! Also, I found a pseudopotential set with atomic orbital included, pseudo-dojo.
I'm now trying to relax TiO2 and hcp-Ti with nc-pp from pseudo-dojo, running 'r2scan' functional.
However, I always observe an energy increase in the relaxation steps. Meanwhile, the stress on the cell is not getting smaller. I've tried different versions of pseudopotentials from pseudo-dojo(v0.4, pbe, pw and pbesol). The problem persists. My input file is at the end of this email. You'll also find output energy and stress after the input file. I'm quite new using a non-pbe functional so I'm not sure which input parameter I should look into.
Thank you so much for your help!
Kefan
INPUT:
&control
calculation='vc-relax',
prefix='TiO2_dojo_relax'
outdir='./output_TiO2_dojo'
pseudo_dir='./dojo04_pw_high'
/
&system
ibrav=6,
celldm(1)=8.532654,
celldm(3)=0.672853,
nat=6,ntyp=2,
ecutwfc=400,
occupations='smearing',
smearing='gaussian',
degauss=0.02,
input_dft='r2scan',
exxdiv_treatment='gygi-baldereschi',
ecutvcut=0.4,
nqx1=4,nqx2=4,nqx3=4
/
&electrons
mixing_beta=0.1,
electron_maxstep=300,
/
&ions
/
&cell
cell_dynamics='bfgs'
/
ATOMIC_SPECIES
Ti 47.867 Ti.upf
O 15.999 O.upf
ATOMIC_POSITIONS (crystal)
Ti 0.500000 0.500000 0.500000
Ti 0.000000 0.000000 0.000000
O 0.200764 0.799236 0.500000
O 0.799236 0.200764 0.500000
O 0.700764 0.700764 0.000000
O 0.299236 0.299236 0.000000
K_POINTS (automatic)
8 8 8 0 0 0
An example of output energy: (This is part of the whole relaxation output, the rest part is similar to what's shown here)
! total energy = -360.68407951 Ry
! total energy = -360.36055154 Ry
! total energy = -360.59563090 Ry
! total energy = -360.66225311 Ry
! total energy = -360.67925379 Ry
! total energy = -360.68326714 Ry
! total energy = -360.68408962 Ry
! total energy = -360.68419620 Ry
! total energy = -360.68412678 Ry
! total energy = -360.68408658 Ry
! total energy = -360.68406230 Ry
Example of output stress:
total stress (Ry/bohr**3) (kbar) P= 9235.88
0.06306395 0.00000000 0.00000000 9277.03 0.00 0.00
0.00000000 0.06306395 0.00000000 0.00 9277.03 0.00
0.00000000 0.00000000 0.06222474 0.00 0.00 9153.58
--
total stress (Ry/bohr**3) (kbar) P= 6003.45
0.04105619 0.00000000 0.00000000 6039.57 0.00 0.00
0.00000000 0.04105619 0.00000000 0.00 6039.57 0.00
0.00000000 0.00000000 0.04031941 0.00 0.00 5931.19
--
total stress (Ry/bohr**3) (kbar) P= 7389.65
0.05050933 0.00000000 0.00000000 7430.18 0.00 0.00
0.00000000 0.05050933 0.00000000 0.00 7430.18 0.00
0.00000000 0.00000000 0.04968288 0.00 0.00 7308.60
--
total stress (Ry/bohr**3) (kbar) P= 8245.72
0.05633717 0.00000000 0.00000000 8287.48 0.00 0.00
0.00000000 0.05633717 0.00000000 0.00 8287.48 0.00
0.00000000 0.00000000 0.05548546 0.00 0.00 8162.19
--
total stress (Ry/bohr**3) (kbar) P= 8723.04
0.05958080 0.00000000 0.00000000 8764.64 0.00 0.00
0.00000000 0.05958080 0.00000000 0.00 8764.64 0.00
0.00000000 0.00000000 0.05873251 0.00 0.00 8639.85
--
total stress (Ry/bohr**3) (kbar) P= 8974.70
0.06129011 0.00000000 0.00000000 9016.09 0.00 0.00
0.00000000 0.06129011 0.00000000 0.00 9016.09 0.00
0.00000000 0.00000000 0.06044601 0.00 0.00 8891.91
--
total stress (Ry/bohr**3) (kbar) P= 9103.60
0.06216601 0.00000000 0.00000000 9144.94 0.00 0.00
0.00000000 0.06216601 0.00000000 0.00 9144.94 0.00
0.00000000 0.00000000 0.06132295 0.00 0.00 9020.92
--
total stress (Ry/bohr**3) (kbar) P= 9170.12
0.06261743 0.00000000 0.00000000 9211.34 0.00 0.00
0.00000000 0.06261743 0.00000000 0.00 9211.34 0.00
0.00000000 0.00000000 0.06177676 0.00 0.00 9087.67
--
total stress (Ry/bohr**3) (kbar) P= 9202.74
0.06283894 0.00000000 0.00000000 9243.93 0.00 0.00
0.00000000 0.06283894 0.00000000 0.00 9243.93 0.00
0.00000000 0.00000000 0.06199889 0.00 0.00 9120.35
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240620/586873aa/attachment.html>
More information about the users
mailing list