The ABINIT Users Mailing List ( CLOSED )

[ilyes.hamdi@fst.rnu.tn]

Dear abinit users
I'm doing an RF-calculation on ZnSe in ZB structure. Comparison with 
experimental data, shows an good agreement of the calculated phonons 
frequencies in IBZ and of the Born-effective charge but the dielectric 
constant is too large (7.52 compared to 6.3). However, the same calculation 
on BN gives for the epsilon infinity  an excellent agreement with experience 
(4.54 compared to 4.50).

I have found in the litterature that this overestimation is in part due to 
the so called band gap problem. (Phys. Rev B54, 11305 (1996).)

Is there another explanation for this discrepancy?, and why it works for some 
compound and not for others.