How can I calculate the ground state energy levels for Fe atom
Posted: Tue Mar 31, 2009 9:23 am
Hi,
I need to calculate the ground state energy levels for a single Fe atom. Following the instructions given in the VASP manual, I first did a non-selfconsistent with the following INCAR:
ISPIN=2
ICHARG=12
MAGMOM=4
Subsequently, I carried out another calculation with fixed orbital occupancies (4s2 3d6) as follows:
ISTART=1
ISPIN=2
MAGMOM=4
ISMEAR= -2
LDIAG=.FALSE.
#ISYM=0
#VOSKOWN=1
AMIX=0.01
BMIX=0.0001
NELM=200
FERWE = 5*1 1*1 10*0
FERDO = 1*1 5*0.2 10*0
I was expecting to see 5-fold degenerated 3d states in the OUTCAR file, however, the produced eigenvalues turned out to be completely distorted as follows:
spin component 1
k-point 1 : 0.0000 0.0000 0.0000
band No. band energies occupation
1 -7.4200 1.00000
2 -7.4147 1.00000
3 -7.2536 1.00000
4 -7.2521 1.00000
5 -6.8955 1.00000
6 -5.0983 1.00000
7 -1.2370 0.00000
8 -1.0703 0.00000
9 -1.0699 0.00000
10 -0.2583 0.00000
11 0.8288 0.00000
12 0.8719 0.00000
13 1.5210 0.00000
14 1.5389 0.00000
15 1.5391 0.00000
16 1.6633 0.00000
spin component 2
k-point 1 : 0.0000 0.0000 0.0000
band No. band energies occupation
1 -4.1917 1.00000
2 -4.0573 0.20000
3 -4.0551 0.20000
4 -3.9186 0.20000
5 -3.6789 0.20000
6 -3.6785 0.20000
7 -0.7007 0.00000
8 -0.7007 0.00000
9 -0.4508 0.00000
10 -0.2153 0.00000
11 0.8479 0.00000
12 0.8802 0.00000
13 1.5783 0.00000
14 1.5784 0.00000
15 1.6405 0.00000
16 1.7234 0.00000
The calculated value of free energy -3.05 is very close to the standard correction energy -3.15 given in the manual. But I'm wondering why my calculated 3d states are not 5-fold degenerate. Does anybody know How I can restore the degeneracy of 3d orbitals?
Your helps are greatly appreciated.
Spak
Additional info: I use PAW potential with PBE functional
I need to calculate the ground state energy levels for a single Fe atom. Following the instructions given in the VASP manual, I first did a non-selfconsistent with the following INCAR:
ISPIN=2
ICHARG=12
MAGMOM=4
Subsequently, I carried out another calculation with fixed orbital occupancies (4s2 3d6) as follows:
ISTART=1
ISPIN=2
MAGMOM=4
ISMEAR= -2
LDIAG=.FALSE.
#ISYM=0
#VOSKOWN=1
AMIX=0.01
BMIX=0.0001
NELM=200
FERWE = 5*1 1*1 10*0
FERDO = 1*1 5*0.2 10*0
I was expecting to see 5-fold degenerated 3d states in the OUTCAR file, however, the produced eigenvalues turned out to be completely distorted as follows:
spin component 1
k-point 1 : 0.0000 0.0000 0.0000
band No. band energies occupation
1 -7.4200 1.00000
2 -7.4147 1.00000
3 -7.2536 1.00000
4 -7.2521 1.00000
5 -6.8955 1.00000
6 -5.0983 1.00000
7 -1.2370 0.00000
8 -1.0703 0.00000
9 -1.0699 0.00000
10 -0.2583 0.00000
11 0.8288 0.00000
12 0.8719 0.00000
13 1.5210 0.00000
14 1.5389 0.00000
15 1.5391 0.00000
16 1.6633 0.00000
spin component 2
k-point 1 : 0.0000 0.0000 0.0000
band No. band energies occupation
1 -4.1917 1.00000
2 -4.0573 0.20000
3 -4.0551 0.20000
4 -3.9186 0.20000
5 -3.6789 0.20000
6 -3.6785 0.20000
7 -0.7007 0.00000
8 -0.7007 0.00000
9 -0.4508 0.00000
10 -0.2153 0.00000
11 0.8479 0.00000
12 0.8802 0.00000
13 1.5783 0.00000
14 1.5784 0.00000
15 1.6405 0.00000
16 1.7234 0.00000
The calculated value of free energy -3.05 is very close to the standard correction energy -3.15 given in the manual. But I'm wondering why my calculated 3d states are not 5-fold degenerate. Does anybody know How I can restore the degeneracy of 3d orbitals?
Your helps are greatly appreciated.
Spak
Additional info: I use PAW potential with PBE functional