Dear all
I used GGA(PBE) with PAW to optimized structure for cubic phase of BaTiO3
why the total energy with respect to volume are not minimized in cubic phase
this is my out put files
etotal1 -1.3310822996E+02
etotal2 -1.3313854779E+02
etotal3 -1.3314459054E+02
etotal4 -1.3314981328E+02
etotal5 -1.3316412103E+02
etotal6 -1.3313097779E+02
etotal7 -1.3314194448E+02
etotal8 -1.3316189991E+02
etotal9 -1.3314829740E+02
etotal10 -1.3315730374E+02
etotal11 -1.3314609766E+02
etotal12 -1.3312207676E+02
etotal13 -1.3315641486E+02
etotal14 -1.3314098530E+02
etotal15 -1.3310855729E+02
I am tired and alone please reply me
This is my input files
# BaTiO3 : computation of the total energy
# pure of Ti.
ndtset 15
acell: 7.00 7.00 7.00
acell+ 0.10 0.10 0.10
#Definition of the basis of the geometry relaxation, the volume of the cell is fixed.
optcell 3
ionmov 2
dilatmx 1.2
ecutsm 0.5
# COMMON INPUT DATA
#Definition of the unit cell
rprim 1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
#Definition of the atom types
ntypat 3
znucl 56 22 8
#Definition of the atoms
natom 5
typat 1 2 3 3 3
xred
0.0 0.0 0.00
0.5 0.5 0.50
0.5 0.5 0.00
0.0 0.5 0.50
0.5 0.0 0.50
#Definition of the planewave basis set
ecut 25.0
pawecutdg 35
pawovlp -2
# find the optimal total energy
ntime 100
tolmxf 1.0d-5
toldff 1.0d-6
#Definition of the k-point grid
kptopt 1
ngkpt 8 8 8
nshiftk 1
shiftk
0.5 0.5 0.5
#Definition of the SCF procedure
nstep 100
#toldfe 1.0d-6
diemac 8
Best.
Abdulmutta
For BaTiO3 in cubic phase does not minimization of total ene
Moderator: bguster
Re: For BaTiO3 in cubic phase does not minimization of total
A few remarks :
1) It looks like the last sensible move from Broyden was step 5. So for next calculation you could start from there.
2) Have you checked that at each SCFcycle step you get a converged result ? Is it the case ?
3) In case you answer yes to 2) Try to improve your k-point sampling and/or your grid parameters or to find references more reliable for your starting structure.
4) In case you answer no to 2) Try to improve the convergence of the SCFcycle through the various ways decribed in the forum and in Abinit documentation.
May I ask you what is the purpose of you loop over acell values ?
regards
1) It looks like the last sensible move from Broyden was step 5. So for next calculation you could start from there.
2) Have you checked that at each SCFcycle step you get a converged result ? Is it the case ?
3) In case you answer yes to 2) Try to improve your k-point sampling and/or your grid parameters or to find references more reliable for your starting structure.
4) In case you answer no to 2) Try to improve the convergence of the SCFcycle through the various ways decribed in the forum and in Abinit documentation.
May I ask you what is the purpose of you loop over acell values ?
regards