BaTiO3 forces wrong using PAW psp.

External optional components of Abinit (BigDFT, Wannier90...) and codes using/providing data from/to Abinit (AtomPAW, ONCVPSP, EXC, DP, Yambo...)

Moderators: ebousquet, bxu

Locked
jiahaoz
Posts: 1
Joined: Fri Apr 30, 2021 7:45 pm

BaTiO3 forces wrong using PAW psp.

Post by jiahaoz » Fri Apr 30, 2021 7:58 pm

Hello everyone,
I was calculating the forces of BaTiO3 under the electrical field. However, I found that the force is not accurate. However, if I switch to Norm Conserving PSP. The results agree so well. I mean F=Q \times E. Any thing wrong with the PAW psp?
I posted my input here:
===================
nstep 50
ecut 25
ecutsm 0.5
pawecutdg 35
pawxcdev 0
occopt 1
prtwf 1
prtden 0
prteig 0
toldfe 1.0d-12

ndtset 2
jdtset 1 2
berryopt1 -1 rfdir1 1 1 1
berryopt2 4 efield2 0.000090692911356079 0.0000900338351967555 0.001058527828369657 getwfk2 1

acell 4.0811309814 4.0810610490 4.0809919294 angstrom
rprim 1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0

#Structural parameters
natom 5 # two atoms in the cell
ntypat 3
nband 20
typat 1 2 3 3 3
znucl 56 22 8
xred 0.00188030 0.00189342 0.01764798
0.51634006 0.51635532 0.52810615
0.97257751 0.48515228 0.48726470
0.48513157 0.97260416 0.48725646
0.48455555 0.48457207 0.97577099
# Kpoint Grid
symmorphi 0
kptopt 1
ngkpt 6 6 6
nshiftk 1 # this Monkhorst-Pack shift pattern is used so that the symmetry of the shifted grid
# is correct. A gamma-centered grid would also have the correct symmetry but would be
# less efficient.
shiftk 1.0 1.0 1.0

pp_dirpath "$ABI_PSPDIR"
# pseudos "pbe_s_sr/Ba.psp8,pbe_s_sr/Ti.psp8,pbe_s_sr/O.psp8"
# pseudos "pbe_s_sr/Ba.psp8,pbe_s_sr/Ti.psp8,pbe_s_sr/O.psp8"
pseudos "ATOMICDATA/Ba.GGA_PBE-JTH.xml,ATOMICDATA/Ti.GGA_PBE-JTH.xml,ATOMICDATA/O.GGA_PBE-JTH.xml"
# pseudos "BTO/Ba.GGA_PBE-JTH.xml,BTO/Ti.GGA_PBE-JTH.xml,BTO/O.GGA_PBE-JTH.xml"
# pseudos "BTO/Ba.xml,BTO/Ti.xml,BTO/O.xml"
# pseudos "BTO/Ba.xml.gz,BTO/Ti.xml.gz,BTO/O.xml.gz"
===================
Best wishes
Jiahao

Locked