My Community

Dear all,

I am trying to use HSE06 and B3LYP to get corrected electronic band gap. I am running single-point on optimized structures but the energies are coming out negative. When I use PBE+D3 or SCAN functionals it runs fine, with reasonable results. Please look at the INCARs and tell me what is wrong about it.
For B3LYP calculations, I set LEXCH=B3 in the POTCAR, as recommended in other post forum (https://www.researchgate.net/post/How_t ... al_in_VASP).

Thank you,
Armando

There are the INCARs.

B3LYP:
SYSTEM = AP B3LYP
LREAL = A
IALGO = 48
SIM = 8
ISYM=3
PREC=Accurate
NPAR = 8
NCORE=8
IVDW=12
LHFCALC = .TRUE.
GGA = B3
AEXX = 0.2
AGGAX = 0.72
AGGAC = 0.81
ALDAC = 0.19
NELM = 300


HSE06:
SYSTEM = HSE06
LREAL = A
IALGO = 48
ISMEAR = 0
NSIM = 1
ISYM=0
ENCUT=1000
PREC=Accurate
NCORE=8
IVDW=12
LHFCALC = .TRUE.
GGA = PE
HFSCREEN = 0.2
NELM = 300

I am using these POTCARS:
PAW_PBE H_GW 21Apr2008
PAW_PBE N_GW_new 19Mar2012
PAW_PBE Cl 17Jan2003
PAW_PBE O 08Apr2002

Hi,

First of all, the POTCAR file should never be modified as written at POTCAR. It is only in INCAR that the functional is specified.

Concerning your INCAR files, they seem ok except for "SIM" for B3LYP which should be a typo (?). And why did you chose NSIM=1 for HSE06? Please, upload the OUTCAR files so that we can have a look at them.

Dear Fabien,

Thank you for your reply.
Please find attached the OUTCARs.

Best regards,
Armando

Hi,

In your first post you mentioned negative energies. Which energies are you referring to? The band gap? According to the orbitals energies in your OUTCAR files, the band gap is positive (4.69 eV with HSE06 and 4.93 eV with B3LYP).

Besides, note the important point: it is currently note possible in VASP to combine HSE06 with DFT-D3 (in your first post the HSE06 INCAR contains "IVDW=12 "). This is something that we have to update in VASP.

Concerning the ISYM tag in INCAR. Its value is 3 for B3LYP, but 0 for HSE06. Be careful to the details.

Finally, I have also noticed that you are using a rather old version of VASP (5.4.4). The recommendation would be to upgrade.