[Wien] mBJ gap GaAs

[tran at theochem.tuwien.ac.at]

Hi,

yes, 1.63 eV is the value that you should obtained with mBJ. This value is
in much better agreement with experiment than LDA or PBE, but you should
not expect perfect agreement with experiment. However, by varying
manually the value of c [Eq. (3) of PRL 102, 226401 (2009)] you can get
more or less any value of the band gap that you want (an increase of
c leads to an increase of the band gap). For this, you have
to specify yourself the values of alpha, beta and e in case.in0abp
(c=alpha and choose 0 and 1 for beta and e, respectively).
You can find some explanations at the very end of Sec. 4.5.9 in the user's
guide.

F. Tran

On Tue, 15 Apr 2014, sollebac wrote:

> Dear wien2k users, 
> Im trying to calculate the gap mBJ  of GaAs as an example following the
> user-guide. Everything finished ok  but the value that  I got is 1.63 eV  at Gamma,  while the experimental value are  ~1.52 (300K)  and ~1.42 (0K).  How can i
> get close value to experimental?  i mean how can i converge the value (1.63)  to get the best value close to experimental?  the k-points are 560 IBZ and is
> not-spin-polarized. 
> thank in advance. 
> Jose Luis              
> 
>