My Community

Dear VASP Developers and Users,

I am currently working on a large system with constrained occupation to calculate the electronic properties using VASP 6.4.3. However, I have encountered an issue in which the energy columns in both the DOSCAR (zeroes in all energy-related columns) and PROCAR (zeroes across all orbital components) files are written as zeros throughout the entire file, even though the EIGENVAL file appears to be written correctly. Excerpts of DOSCAR and PROCAR files are:
DOSCAR:
3.416 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
3.417 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
3.419 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
3.420 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
3.421 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
3.423 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00

PROCAR:
ion s py pz px dxy dyz dz2 dxz x2-y2 tot
1 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
5 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

INCAR
SYSTEM = AxB1-x
ALGO = Fast
LREAL = A
ENCUT = 400
LCHARG = .TRUE.
EDIFF = 1E-6
ISMEAR = -2
FERWE = 3582*1.0d0 898*0.0d0
# Ionic Relaxation
NSW = 0
IBRION = -1
NCORE = 32
NBANDS = 4480
NELM = 200
NELMIN = 4
AMIN = 0.01
ICHARG = 11
LORBIT = 11
NEDOS = 6000
EMAX = 7
EMIN = 3

For reference, I also reproduced the same issue with another system (Si).
I would greatly appreciate any insights into what might be causing this problem with the DOSCAR and PROCAR files. Is there a parameter in the INCAR file that could be affecting the correct writing of the DOSCAR and PROCAR files or is there a potential bug or issue with VASP 6.4.3?

Thank you very much in advance.
Saif

Saif Ullah
Research Assistant Professor
Wake Forest University
Winston-Salem, NC, USA

Hi Saif,

Thank you for you question. Could you provide the POSCAR, POTCAR, KPOINTS, and INCAR files and I'll see if I can reproduce it.

Best wishes,

Chris

Dear Chris,

Thanks for the response. The input files for 216 Si atoms are attached.
Please let me know if you need anything further.

I appreciate your help!

Thanks,
Saif

Dear Saif,

Thank you. I will try to repeat it now and get back to you.

Best wishes,

Chris

Dear Saif,

I have managed to repeat your calculation and see the same issue. It is not clear to me exactly why this is yet but I will investigate further and get back to you.

Edit:
It seems that the DOS was never intended to be used for constrained occupation. The DOS is generated from the smearing, so if ISMEAR=-2 is set, then it does not enter the routine that generates the DOS and so there is nothing output to DOSCAR, hence the zeroes.

Is there any particular reason why you want to use constrained occupation? Perhaps there might be an alternative possible.

Best wishes,

Chris

Edit: added comments on implementation

Dear Chris,

Thank you very much for looking into that. I appreciate your efforts.

Our real system is much more complicated than the Si. We want to study charged systems with fixed partial occupancies for which we need constrained occupation.

best,
Saif

Dear Saif,

I see. Perhaps you could use constrained occupation to fix the electronic structure, then try doing a single elementary step with different smearing (e.g. Gaussian ISMEAR = 0 and SIGMA = 0.05) and see what the DOS looks like from that. It should be relatively unchanged but may be meaningful. It worked for me for your Si, although the structure looked quite odd, so I think that there is room to improve it.

Best wishes,

Chris