[VASP 5.2.11] bug or feature? PDOS for WC

Problems running VASP: crashes, internal errors, "wrong" results.


Moderators: Global Moderator, Moderator

Locked
Message
Author
maksim

[VASP 5.2.11] bug or feature? PDOS for WC

#1 Post by maksim » Mon Feb 27, 2012 9:29 am

Dear developers,

I did calculation of projected DOS for a hexagonal WC. I can see that there is a very strong peak appearing above the Fermi level in spd-projected DOS for W. When I integrated the area under the peak, I got totally 16 electrons. All these electrons appear in integrated s-DOS of W. Is my guess that these are combined 6s and 5f states of W correct? Why they are not in separated columns, say, s-DOS and f-DOS? Would appreciate any comments of this issue.

POSCAR:
C W # WC
1.0
2.5167 -1.453 1.7794E-016
0 2.906 1.7794E-016
0 0 2.837
1 1
Direct
0.333333 0.666667 0.5
0 0 0

KPOINTS:
not only gamma point
0
Gamma
6 6 8
0 0 0

POTCARs:
POTCAR: PAW_PBE C 08Apr2002
POTCAR: PAW_PBE W_pv 06Sep2000

INCAR:
System = WC

ISTART = 0
ICHARG = 2

EDIFF = 1E-06 # stopping-criterion for ELM

EDIFFG = -1E-02
NSW = 0 # number of steps for IOM
IBRION = -1 # ionic relax: 0-MD 1-quasi-New 2-CG
ISMEAR = 0
SIGMA = 0.07
EMIN = -40
EMAX = 40
NEDOS = 8000

ISIF = 2 ! 3 - stress and relaxation; 2 - relax ions

LORBIT = 11
LVTOT = F # local potential
LWAVE = F
LCHARG = F
LEFL = F

POTIM = 0.2

PREC = Normal

ALGO = Fast
NSIM = 4

LREAL = F

SYMPREC=1e-6

<span class='smallblacktext'>[ Edited ]</span>
Last edited by maksim on Mon Feb 27, 2012 9:29 am, edited 1 time in total.

maksim

[VASP 5.2.11] bug or feature? PDOS for WC

#2 Post by maksim » Thu Mar 22, 2012 3:33 pm

Ok, we've found what the problem was. The pseudopotential W_pv_new appears to be broken, there are no ghost peaks observed when the pseudopotelntials W and W_pv are used. Thank you for your kind attention.

Max
Last edited by maksim on Thu Mar 22, 2012 3:33 pm, edited 1 time in total.

Locked