DOS with a strange band

[italop]

Hi guys,

I'm doing some DOS calculations but when i try to calculate the DOS for CO (yes, just carbon monoxide), im getting a peak at 0.0 which i can not explain. When I put the total DOS and the projected DOS, there is not even one contribution to that peak. How am I supposed to explain that?

When I get the integration of the bands, excluding the 0.0 peak, I get almost 14 electrons (which is expected or at least I think so...). Vasp compute the core electrons in DOS, right?

POSCAR, POTCAR and KPOINTS files are OK. The INCAR is shown below:

--- INCAR FILE ---

ISTART = 0 # startjob: no WAVECAR file
SYSTEM= Zn O
ICHARG = 2 # charge: from atoms
NELM = 100 # maximum of 40 electronic steps
NELMIN = 12 # minimum of two steps
NELMDL = -5 # no update of charge for 3 steps
EDIFF = 10E-6 # accuracy for electronic minimization
ENCUT=400
IBRION=-1
NSW=0
ISIF=2
LREAL=.FALSE.
GGA=91
PREC=Accurate
ISYM=0
LCOMPAT=.TRUE.
LORBIT=12
NEDOS=1000
ISMEAR=-1
EMIN=35
EMAX=-35

--- END OF INCAR FILE ---

I have tried lots of things, but didnt work. In the other bands I get projected orbitals of oxygen and carbon just fine.

Total DOS image:


Thanks in advance.

[admin]

1) free molecules do not show a DOS, but a discrete spectrum
2) if you want to plot the DOS of CO@(some surface) the peaks related to CO may be very narrow. Therefore
--> have a look in PROCAR, at which the KS states the CO-contributions are found
--> check the dispersion of these peaks
--> set the density of the energies for which the DOS is evaluated such that these peaks can be resolved. This can be done either by setting NEDOS (default 301), by choosing a smaller range of energies for which the DOS is calculated (EMIN,EMAX), or both.