I am attempting to understand the state-projected DOS written to the DOSCAR (e.g., when using LORBIT=11) and the total charge from the OUTCAR.
I am looking at Ru (hcp structure, 2 atoms in unit cell) with three different PBE-pseudopotentials: Ru, Ru_pv, Ru_sv
Here is my INCAR:
ENCUT=520
PREC=Accurate
EDIFF=1.e-7
EDIFFG=1.e-5
LREAL=Auto
ROPT=1.e-4
ISMEAR=-5
ISPIN=2
LORBIT=11
MAGMOM= 2*1.
KSPACING=0.1
KGAMMA=.TRUE.
NEDOS=9999
When I look at the total charge in the OUTCAR, I get
Ru:
# of ion s p d tot
------------------------------------------
1 0.618 0.775 6.546 7.939
2 0.618 0.775 6.546 7.939
--------------------------------------------------
tot 1.236 1.550 13.091 15.877
Ru_pv
# of ion s p d tot
------------------------------------------
1 0.394 6.295 5.919 12.608
2 0.394 6.295 5.919 12.608
--------------------------------------------------
tot 0.788 12.590 11.838 25.216
Ru_sv
# of ion s p d tot
------------------------------------------
1 2.216 6.097 5.276 13.589
2 2.216 6.097 5.276 13.589
--------------------------------------------------
tot 4.433 12.193 10.552 27.178
I do understand the case of Ru. However, I do not understand why adding the 3p-states reduces the occupation of s- and d-states, and why for Ru_sv the d-state occupation reduces further. In addition, the total charge agrees with the total number of electrons for Ru, but why does it not when I add the low-lying s- and p-states?
Also, the total DOS-plot near the Fermi energy agrees almost perfectly for these three potentials. However, when I plot the partial DOS from the DOSCAR, I get the attached plot for the s-states. The three curves agree very well in shape, but the area becomes smaller for Ru_pv and even smaller for Ru_sv.
I do understand that projected states need not add up to the number of electrons in general (because electrons in the conduction band may not overlap well with atomic states), but I do not understand why the addition of low-lying states causes this change in the charges and the DOS level, although the shape of the DOS remains almost unchanged. (As does the total DOS which looks almost exactly the same for the three potentials.) Any help would be appreciated.
Dear Mrtin
i think that you should consider the nature and the quality of the used pseudoptentials. which states are included in this PP during the generations step?
which states are valence states? semicore states are included in the PP or not?
the PP is scalar relativistic or not? all these parameters can affect the DOS curves and the partial charges?
best regards
Hi,
Step by step process of accurate (both total and projected) DOS and band structure calculations using VASP and some ways of plotting the band structure and DOS of our system is explained in a Youtube video.
The link is:
https://www.youtube.com/watch?v=RYziHtNFaPs
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Atomic, Molecular and Optical Physics
- Atoms