QuantumATK Forum
QuantumATK => General Questions and Answers => Topic started by: bfazi on December 7, 2016, 15:37
-
Hello,
I have calculated the bandstructure of a germanium nanowire and I attached my input script in the following. I can't understand why the Fermil level of the system is near the valence band!!! I did not add any doping atom in the structure.
Please tell me how can I solve this issue?
Thanks
-
Where should it be? The Fermi energy lies in the middle of the gap only for bulk non-doped semiconductors.
I also see you are using a semiempirical method to describe your structure. Are you sure that the parametrization works well for your system? Did you check it against DFT ?
Regards,
Daniele.
-
Hello Azi,
I looked a little bit into the problem. The main issue there is that the band gap calculated using the Huckel method is very large (around 5 eV), whereas the band gap calculated using DFT is much smaller (around 1 eV).
From a numerical point of view, at given temperature (in your calculations, the electronic temperature is k_B*T = 0.025 eV) it becomes increasingly difficult to resolve the position of the Fermi level in the gap when the energy gap increases. So, if you want the Fermi level positioned correctly in the nanowire calculated using Huckel, you have to increase the temperature.
Regards,
Daniele.
-
The Muller/Hoffman Huckel parameters are really not designed to give correct band gaps for Ge, or Si for that matter. Cerda has refitted the parameters for Si, so there you can get a good gap, but no such parameter set exists for Ge.
I would suggest using Slater-Koster tight-binding e.g. the Bassani parameters or Boykin, if you want to study Ge nanowires.
-
Yes, provided you set up the calculations correctly (in particular, as a self-consistent one).
-
The shape of the IV is a bit odd, I admit, but since you add doping and increase the temperature quite a bit, I don't see anything wrong in the current onset being so low.
-
Great! Thanks for reporting :)