Structure Optimization of PbTiO3
Posted: Mon Mar 02, 2009 1:34 am
Dear Vasp Users and administrators
I am a beginner of vasp. I'm now calculating the spontanous polarization of PbTiO3. So I have to optimize the atomic positions at first. I have used two kinds of methods to optimize the atomic positions, i.e ionic relaxation and MD simulation. Unfortunately, there are big difference between my optimized results of the two kinds methods and the optimized results of okuno who is a vasp user as well. The following are my INCAR files and optimized results ot the two methods.
I. ionic relaxation:
INCAR:
SYSTEM=PTOTetragonal
ENCUT=650
ISTART=0
ICHARG=2
ISMEAR=0
SIGMA=0.04
NSW=10
NELMIN=5
IBRION=2
ISIF=2
POTIM=0.1
EDIFF=1E-6
EDIFFG=-1E-5
PREC=ACCURATE
the optimized result of ionic relaxation:
Tetragonal PTO
1.00000000000000
3.8958188134500000 0.0000000000000000 0.0000000000000000
0.0000000000000000 3.8958188134500000 0.0000000000000000
0.0000000000000000 0.0000000000000000 4.1685261303900001
1 1 3
Direct
0.0000000000000000 0.0000000000000000 0.0107242005892228
0.5000000000000000 0.5000000000000000 0.5194356555706965
0.5000000000000000 0.5000000000000000 -0.0268471007195760
0.5000000000000000 0.0000000000000000 0.5008436222798296
0.0000000000000000 0.5000000000000000 0.5008436222798296
II. MD simulation
INCAR:
SYSTEM=PTO Tetra
#ENCUT=400
#ISTART=0
#ICHAGR=2
ISMEAR=0
SIGMA=0.1
IBRION=0
NSW=10000
POTIM=1.0
#EDIFF=1E-5
PREC=Accurate
SMASS = -3
TEBEG = 295
TEEND = 295
the optimized result of MD simulation:
Tetra PTO
1.00000000000000
3.8958188134500000 0.0000000000000000 0.0000000000000000
0.0000000000000000 3.8958188134500000 0.0000000000000000
0.0000000000000000 0.0000000000000000 4.1685261303900001
1 1 3
Selective dynamics
Direct
0.0000000000000000 0.0000000000000000 0.0000000000000000 F F F
0.5000000000000000 0.5000000000000000 0.4178780680825455 F F T
0.0000000000000000 0.5000000000000000 0.4646093864408842 F F T
0.5000000000000000 0.0000000000000000 0.4490985865019561 F F T
0.5000000000000000 0.5000000000000000 0.9772940646253278 F F T
And the optimized result of okuno is:
PbTiO3-tetra
1.00000000000000
3.8958188134500000 0.0000000000000000 0.0000000000000000
0.0000000000000000 3.8958188134500000 0.0000000000000000
0.0000000000000000 0.0000000000000000 4.1685261303900001
1 1 3
Direct
0.0000000000000000 0.0000000000000000 0.0000000000000000
0.5000000000000000 0.5000000000000000 0.5380810847389981
0.5000000000000000 0.5000000000000000 0.1044018384179992
0.0000000000000000 0.5000000000000000 0.6165899713899989
0.5000000000000000 0.0000000000000000 0.6165899713899989
I have confirmed that the optimized result of okuno agreed well with the experimental data. So what' s wrong with my methods? I am looking forward to suggestion of everyone.
Thank you very much!
I am a beginner of vasp. I'm now calculating the spontanous polarization of PbTiO3. So I have to optimize the atomic positions at first. I have used two kinds of methods to optimize the atomic positions, i.e ionic relaxation and MD simulation. Unfortunately, there are big difference between my optimized results of the two kinds methods and the optimized results of okuno who is a vasp user as well. The following are my INCAR files and optimized results ot the two methods.
I. ionic relaxation:
INCAR:
SYSTEM=PTOTetragonal
ENCUT=650
ISTART=0
ICHARG=2
ISMEAR=0
SIGMA=0.04
NSW=10
NELMIN=5
IBRION=2
ISIF=2
POTIM=0.1
EDIFF=1E-6
EDIFFG=-1E-5
PREC=ACCURATE
the optimized result of ionic relaxation:
Tetragonal PTO
1.00000000000000
3.8958188134500000 0.0000000000000000 0.0000000000000000
0.0000000000000000 3.8958188134500000 0.0000000000000000
0.0000000000000000 0.0000000000000000 4.1685261303900001
1 1 3
Direct
0.0000000000000000 0.0000000000000000 0.0107242005892228
0.5000000000000000 0.5000000000000000 0.5194356555706965
0.5000000000000000 0.5000000000000000 -0.0268471007195760
0.5000000000000000 0.0000000000000000 0.5008436222798296
0.0000000000000000 0.5000000000000000 0.5008436222798296
II. MD simulation
INCAR:
SYSTEM=PTO Tetra
#ENCUT=400
#ISTART=0
#ICHAGR=2
ISMEAR=0
SIGMA=0.1
IBRION=0
NSW=10000
POTIM=1.0
#EDIFF=1E-5
PREC=Accurate
SMASS = -3
TEBEG = 295
TEEND = 295
the optimized result of MD simulation:
Tetra PTO
1.00000000000000
3.8958188134500000 0.0000000000000000 0.0000000000000000
0.0000000000000000 3.8958188134500000 0.0000000000000000
0.0000000000000000 0.0000000000000000 4.1685261303900001
1 1 3
Selective dynamics
Direct
0.0000000000000000 0.0000000000000000 0.0000000000000000 F F F
0.5000000000000000 0.5000000000000000 0.4178780680825455 F F T
0.0000000000000000 0.5000000000000000 0.4646093864408842 F F T
0.5000000000000000 0.0000000000000000 0.4490985865019561 F F T
0.5000000000000000 0.5000000000000000 0.9772940646253278 F F T
And the optimized result of okuno is:
PbTiO3-tetra
1.00000000000000
3.8958188134500000 0.0000000000000000 0.0000000000000000
0.0000000000000000 3.8958188134500000 0.0000000000000000
0.0000000000000000 0.0000000000000000 4.1685261303900001
1 1 3
Direct
0.0000000000000000 0.0000000000000000 0.0000000000000000
0.5000000000000000 0.5000000000000000 0.5380810847389981
0.5000000000000000 0.5000000000000000 0.1044018384179992
0.0000000000000000 0.5000000000000000 0.6165899713899989
0.5000000000000000 0.0000000000000000 0.6165899713899989
I have confirmed that the optimized result of okuno agreed well with the experimental data. So what' s wrong with my methods? I am looking forward to suggestion of everyone.
Thank you very much!