[Pw_forum] EXC (hybride) functionals

[Eyvaz Isaev]

Dear Stefano,
Dear Paolo,
Dear developers,

I would like to deal with semiconductors doped by
transitions metals. It is well known that in this case
the main  problem comes from the band-gap problem.
Sure, I can use the LDA+U approximation, but it looks
somewhat tricky, but not full ab initio. May be I am
wrong on this issue, sorry. 
 
I tried to learn more about on this problem and found
that there are some kind of EXC (hybride) functionals
which give good results for semiconductors' band-gap.
Besides, energy bands could be calculated for any
k-point in contrast to GW-part of ABINIT. This kind of
functionals are included in Gaussian03 and CRYSTAL03
packages which operate with atomic orbitals as basis
functions (at least, for the last program). 
 
I have found some controversial issue on this problem.
 In  Journ.Chem.Phys.,v.112,p.1670 (A.Daniel Boese
et.al.) the authors  noticed that "... they
("Car-Parrinello simulations"  in the text) are
precluded from including  exact exchange, because of
PW basis ...", but in VASP  homepage there is an
announcemnent about VASP 5.0, where there are going to
implement PBE0 functional. Even I saw Kresse's
presentation about this one. 
 
So, I would like to ask you, are we really precluded
from implementaton of this kind of EXC (hybride)
functionals in PWSCF?
 
Best regards,
Eyvaz.




 


		
__________________________________ 
Do you Yahoo!? 
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/