[Pw_forum] metallic or not
proffess at yandex.ru
Tue Apr 18 23:15:58 CEST 2006
Thanks a lot for the reply. Yes, the underestimation was the first thing I thought. I use PBE functional and test for silicon diamond gave me a band gap of ~0.8 eV, which is the _standart_ DFT band gap value.
Another thing is that when this clathrate was relaxed by TBMD, the final structure is semiconductor with a band gap around 1.6 eV which is appropriate value. When I relaxed this "relaxed by TBMD" structure it becames a metallic. When I compared two DOS (one from TBMD and another one from PW) I see that are almost identical with one difference. The DOS obtained by PW conatains some additional peaks on Fermi Level.
I think this is a DFT fault. I remember, that in case of carbon clathrates people using DFT found that these structures are metallic, which WAS not confirmed by experiments.
More information about the users