please help me!
I am in "desperate" situation....
I am optimizing cubic Ge (8 atoms cell). according to Monkhorst Pack sampling 10x10x10.
Once I finish the spin polarized optimization I observe that bands are all singly occupied.
E-fermi : 3.0029 XC(G=0): -10.3944 alpha+bet : -8.6014
add alpha+bet to get absolut eigen values
....
.....
55 2.9173 1.00000
56 2.9173 1.00000
57 3.6096 0.00000
58 3.6096 0.00000
....
And this occupation holds for all the k-points (56th band filled 57th empty). But I know that for bandcalculations I need to make a non self consistent calculation. Additionally I would like to add LDA+U correction to the gap.
Then I restart my calculation according this INCAR
ISTART=0
ISMEAR=-1
SIGMA=0.0002
LReal=False
PREC=High
EDIFF=1E-5
EDIFFG=-0.01
LDAUPRINT=2
ISPIN=2
IBRION=2
LDAU=.TRUE.
LDAUTYPE=2
LDAUL=1
AMIX = 0.2
BMIX = 0.0001 ! almost zero, but 0 will crash some versions
AMIX_MAG = 0.8
BMIX_MAG = 0.0001 ! almost zero, but 0 will crash some versions
LDAUU=2.0
LDAUJ=0.0
ENCUT=287.594
ENAUG=531.356
I am applying an onsite Culomb correction on the p orbitals of Ge.
What I get in my OUTCAR is still that the 56th band is filled and the 57th empty.....So everything seems to move towards the right direction, when I do the ICHARG=11 calculation according I get
E-fermi : 3.2090
k-point 1 : 0.0000 0.0000 0.0000
band No. band energies occupation
...
....
52 0.1056 1.00000
53 0.1056 1.00000
54 3.0132 1.00000
55 3.2929 0.00000
56 3.2929 0.00000
BUT (!)
k-point 4 : 0.0600 0.0000 0.0600
band No. band energies occupation
....
...
53 0.1327 1.00000
54 2.2584 1.00000
55 3.1329 1.00000
56 3.2725 0.00000
57 3.6729 0.00000
So the 55th band is now filled. And this even if I apply the LDA+U correction. What is the meaning of this?Must I conclude that Ge remains metallic even with the Hubbard correction? Where am I failing? I have tried ALL the possible combination but when I plot the bands I "see" a metallic system with the Fermi level crossed by this 55th band
and later by the 56th ....Indeed
k-point 9 : 0.1600 0.0000 0.1600
band No. band energies occupation
...
...
55 2.3506 1.00000
56 3.1419 1.00000
57 3.8648 0.00000
58 4.1271 0.00000
....
....
please help me!
any suggestion for improving the calculations?
G
