[Pw_forum] why I can not get the correct band gap of TiO2?

Eyvaz Isaev eyvaz_isaev at yahoo.com
Wed Oct 26 14:38:29 CEST 2005


Hi, 
 
First, some technical comments.
> 
>  ecutwfc=16,

It looks very small, use 30Ry. Besides add ecutrho,
say, 300Ry, for  the augmented charge.

> 
> &IONS
> 
>   ion_dynamics='bfgs',
> 
>   upscale=10
> 
>   /

For the band gap problem you do not need ionic
dynamics.

 
> 
> K_POINTS 
> 
> 36
K-points you used looks strange. How did you generate
these points? They have equal weight and seem to be
k-points along high symmetry directions.
Use K_POINTS {automatic}.
 
Now about your question. DFT is known to give
underestimated and sometime wrong band gap for
semiconductors and insulators. You can improve it
using the LDA+U approximation.   
As far as I am aware Exact Exchange is still in
development in Espresso. 

Bests,
Eyvaz.


		
__________________________________ 
Yahoo! FareChase: Search multiple travel sites in one click.
http://farechase.yahoo.com



More information about the users mailing list