I'd say you'd only need ghost atoms (the basis functions you talk about) if
you need to calculate the formation energy of the vacancy - the so-called
BSSE correction.

Marcos

On Wed, Apr 21, 2010 at 7:49 PM, Shelton Jr, William Allison <
[email protected]> wrote:

> I might be mistaken but doesn't siesta use atom-centered basis ?  If so,
> don't you need to put some basis functions on the vacant site ?
>
> Cheers
> Bill
>
>
> On 4/21/10 1:39 PM, "Marcos Veríssimo Alves" <
> [email protected]> wrote:
>
> Just as a vacancy in real life, a vacancy in a DFT calculation is not an
> addition, but a removal... :)
>
> All you need to do is to remove the undesirable atom(s) from the atom
> coordinates list in your supercell. Of course, don't forget to update the
> number of atoms.
>
> Marcos
>
> On Wed, Apr 21, 2010 at 7:30 PM, Shelton Jr, William Allison <
> [email protected]> wrote:
> Can siesta treat vacancies ?  If so, how can it be added to the input file
> ?
>
> Cheers
> Bill
>
>
>

Responder a