Q about E-fermi and alpha+bet

[des]

Hello everyone,
Is there any who can tell me the meaning of the three terms?
E-fermi : 1.1728 XC(G=0): -5.7140 alpha+bet : -6.3141

I know the first one but it says the term alpha+bet should be added to the E-fermi get the absolute E-Fermi . What I don't understand is which E-fermi should be taken to say something about the fermi level of a particular system?

Thank you

[cpp6f]

The fermi level (and the eigenvalues) are calculated based on a potential that does not include the G=0 component of the pseudopotential or the G=0 component of the coulombic potential (which is ill defined anyway). The term XC(G=0) is the G=0 component of the exchange-correlation potential. The alpha part of Alpha+bet is the G=0 term of the pseudopotential * (ZVAL/NELECT) where ZVAL is the total valance of the ions and NELECT is the total number of electrons (equal in a neutral cell). I have no idea what the beta part of Alpha+bet is.

In my opinion, Alpha+bet is useless for calculating the work function or correcting the eigenvalues to be relative to vacuum. The best way to do this for a slab calculation is to use IDIPOL=3 and LDIPOL=.TRUE. (if the surface is normal to the z direction). This will correct for any dipole interactions between neighboring slabs and will also give you the vacuum potential on both sides of the slab. Just subtract the vacuum potential for the side of the slab that represents the actual surface from the fermi level to get the work function.

[admin]

if your vacuum size is thick enough, the potential should level off to a constant value in the center of the vacuum. Please use the LVTOT=True input parameter and use the tool vtotav.f (with the LOCPOT file) to get the "zero" vacuum potential. The Fermi level should be taken as difference between the calculated fermi energy and this vacuum level then.

[des]

Hi,
Where do I get the tool vtotav.f ? I do not see it anywhere in the web.
Thank you

[WolverBean]

Six years too late, but I can answer the last question: send an email to materialphysik@univie.ac.at and request a copy of vtotav.f.