[Pw_forum] Charged impurity

[mkondrin]

Thank you very much for your reply! Your example is a decisive argument.

My system  is a color center in diamond formed by semiconducting 
impurity and indeed it very slightly influences the Fermi level of the 
system so you are right in this regard too.

Sincerely yours,

M. V. Kondrin

Giuseppe Mattioli wrote:
> Dear M.V.Kondrin
> An impurity or a dopant in a host matrix cannot be considered as separated from the electronic structure of the matrix. Let us suppose that you have a 
> substitutional Mn atom which replace a Ga atom in a GaAs supercell, a very well investigated example. If the cell is neutral then you must have a Mn 
> 3+ with 4 unpaired electrons. A fifth Mn 3d majority orbital, mixed with some contribution with the neighboring As orbitals, will be unoccupied: an 
> acceptor level slightly above the GaAs valence band. If you put an excess electron in the system, then such an electron fills the Mn orbital which is 
> very happy to restore a Mn 2+ state with five unpaired majority electrons. Local changes in the Mn electronic environment are in this case quite 
> significant. On the left hand, if you subtract one electron from the same system, you necessarily subtract it from the GaAs valence band maximum, which 
> is basically spread across all the As 2p orbitals, so that changes in the local electronic structure of all the atoms of this charged supercell are 
> negligible. Differences such as those sketched above generally reflect the interesting Physics of doped semiconductors, and often depend also on the 
> level of theory used to describe their electronic properties (GGA, DFT+U, EXX, ...). As I do not know your system, I can only say that your impurity 
> seems to be "passivated", that is, it does not react to changes of the Fermi level of the system.
> HTH
> Giuseppe
>
>