ENCUT convergence in absorption spectra

[sponza]

Dear users and admin,
I am using VASP 5.3.3 to compute the response function of ZnO. I am computing the optical absorption (RPA without local fields) at the PBE level for different polymorphs of ZnO.

For small cells (zincblende = 1 unit formula, wurtzite = 2 unit formulas), I don't observe any alarming behaviour. PBE Spectra are perfectly converged with ENCUT=500.00 eV.

For bigger structures (sodalite = 6 unit formulas, cubane = 8 unit formulas) only the low energy part of the spectrum (up to ~5 eV) is at convergence for 500.00 eV, above, it seems to be necessary a much higher cuoff and even at ECUT=1400.00 eV the spectrum is not converged. Please note that the structures are similar. Moreover the convergence trend is not monotonic: in the case of cubane (8 unit formulas) the spectrum obtained with ENCUT=900.00 is completely different from encompassing values.

I used "PAW Zn_GW 09Oct2010" pseudopotential for Zn and " PAW O_GW 19Mar2012" pseudopotential for O. The K-point grid is at convergence.

Here the INCAR I used:
ENCUT = 900.00 eV
EDIFF = 0.1E-4 # when stops the Selfconsistent cycle
NPAR = 16
NBANDS = 300
PREC = Normal # Normal precision (influence FFT)
LREAL = .FALSE. # real space projection | .FALSE. or Auto
LCHARG = TRUE
LWAVE = TRUE
LOPTICS = TRUE ; LRPA = TRUE
CSHIFT = 0.1
NEDOS = 2001 ; OMEGAMAX = 40.00 eV
ISMEAR = 0
SIGMA = 0.05
MAXMEM = 4000

I will really appreciate any comment or suggestion.
Thanks!

[support_vasp]

Hi,

We're sorry that we didn’t answer your question. This does not live up to the quality of support that we aim to provide. The team has since expanded. If we can still help with your problem, please ask again in a new post, linking to this one, and we will answer as quickly as possible.

Best wishes,

VASP