With a cluster it is not necessary to use k-points, one just needs the cell
to be sufficiently large to make interactions between images neglectable.

Marcos

On Tue, Apr 20, 2010 at 11:20 AM, Pablo A. Denis <[email protected]> wrote:

> Hi Daniel,
>
>                   If it is a cluster (no periodic conditions), is it
> necessary to use use plenty of kpoints?
>
> Regards,
>
>                   Pablo
>
> ----- Original Message ----- From: "Backlund, Daniel" <
> [email protected]>
> To: <[email protected]>
> Sent: Tuesday, April 20, 2010 4:00 PM
> Subject: RE: [SIESTA-L] Is the siesta suitable for a cluster of metal
>
>
>
>
> SIESTA is capable of your (computationally intensive) problem. When working
> with metals,
> be sure to use a large MeshCutoff ( > 350 Ry.) as well of plenty of
> k-points. If you have a lot of
> k-points, then you can use the Diag.ParallelOverK option.
>
> Daniel J. Backlund
> Texas Tech University - Physics
> [email protected]
> ________________________________________
> From: Bin Shao [[email protected]]
> Sent: Tuesday, April 20, 2010 8:03 AM
> To: siesta-l
> Subject: [SIESTA-L] Is the siesta suitable for a cluster of metal
>
> Dear all,
>
> I intend to calculate the magnetic property of a cluster of metal with
> about one thousand atoms implanted a isolated Fe impurity. But I have no
> idea whether the siesta is suitable for this metallic system. Please give me
> some suggestions about this.
>
> Thank you in advance.
>
> --
> Bin Shao
> College of Information Technical Science, Nankai University
> 94 Weijin Rd. Nankai Dist. Tianjin 300071, China
> Email: [email protected]<mailto:[email protected]>
>


Responder a