The ABINIT Users Mailing List ( CLOSED )

[hwlalves@funrei.br]

As Thomas had commented, I would like to put mine:
In the talk of F. Bechstedt(From Jena) in the IWN2002 held at Aachen last 
july, which attempted to the gap problem of InN(in both wurtzite and 
zincblende structure) we could conclude the following - Probably there is a
"bug" of LDA in dealing the p-d(p from the conduction band and the 4d valence 
elctrons) interaction on band structure calculations. In InN it seems to be 
such the case. Prof. Bechstedt solved this problem not using smearing effects 
(as suggested by Fiorentini) but using the GW correction once recent 
experimental results(by Davydov's group) shows that the band gap of InN in 
the wurtzite structure is around 0.7 eV(showing that the system is really 
semiconductor). I recently did calculations for all nitrides(AlN, GaN and 
InN) using both Troullier-Martins(following Fuchs recipe) and HGH 
pseudopotentials 
in the particular structures - wurtzite, zincblende and rocksalt(I am 
interested in obtain the phonon dispersions of the bulk and of some surfaces) 
and the results agree with Bechstedt ones. In both wurtzite and zincblende 
structures we found zero gap for InN with both pseudopotentials! In the 
rocksalt one, I tested some hints from Bechstedt once this structure results 
from applied [100] pressure over zincblende one(TA(X) phonon-assisted second 
order phase transition), and the gap appears!!! So, if your calculations 
involve gap states I advise you to include GW corrections, if not, the abinit 
results will agree very well with the experimental data. The Bechstedt's 
results will appear soon in the first issue of physica status solidi (c) 
which will be the Proceedings of the International Workshop of Nitrides 
Semiconductors (IWN2002) that I have mentioned above.

Greetings,

Horacio W. Leite Alves