My Community

Posted: **Tue Aug 05, 2014 12:24 pm**

Hi,
anyone has experience in GW calculations of metals?
We have tried to calculate G0W0 corrections of Al, but obtained a strange result that the band is widened, which is unphysical.

This is INCAR of the GW step:

ENCUT=400
NBANDS = 120
ISMEAR = -5
NELM = 1
ALGO = GW0
NOMEGA = 200
NOMEGAR = 200
LSPECTRAL = .TRUE.
LOPTICS = .TRUE.

Posted: **Wed Aug 06, 2014 12:54 pm**

please read the manual: for metals, LOPTICS should NOT be set to true, WAVEDER should not be used.

Posted: **Fri Aug 08, 2014 10:06 pm**

[quote="admin"]please read the manual: for metals, LOPTICS should NOT be set to true, WAVEDER should not be used.
[/quote]
Excuse me, in the manual it suggests that LOPTICS = false in scGW. So it means this is mandatory even for G0W0?

I tried to set LOPTICS=false in the DFT/GW parts, but the problem still exists.