[Pw_forum] Ge band-gap

[Paolo Giannozzi]

On Oct 15, 2007, at 13:47 , Helen wrote:

> I have tried calculating the band-gap of Ge in a similar way to  
> example01
> for Si, and obtained a direct band-gap=0.75, indirect band-gap=0.46  
> at a
> lattice constant=10.614 (optimized lattice constant, converged   
> ecutwfc).
> At the experimental lattice constant the direct band-gap=0.51,  
> indirect
> band-gap=0.38. In literature the band-gaps for LDA are quoted as being
> negative.

the exact value of the band gap in Ge (in GaAs as well) is sensitive to
everything: the lattice parameter, the pseudopotential, the spin-orbit
splitting, the phase of the moon,... If I remember correctly, all- 
electron
LDA results give zero direct gap. Then you have the PP approximation;
moreover norm-conserving PPs are converted to the "separable" form
by choosing a l channel ("local") and transforming the difference  
between
the other channels and the local one into a sum of projectors.  
Results are
very close to, but not exactly equal to, those for the original  
("semilocal")
form, and they depend upon the choice of the local part. So unless you
have evidence that different codes yield different results for exactly
the same system, I don't see any reason to doubt about your results.

Paolo
---
Paolo Giannozzi, Dept of Physics, University of Udine
via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222