Dear Andrei,
You can use the new feature in Siesta-3.0 that initializes the SCF
cycle using a previously generated RHO or DRHO file (in netCDF
format). One diagonalization will then give you an approximation to
the DM in the new basis. Or you could let the SCF cycle finish.
Best regards,
Alberto
On Sun, Nov 29, 2009 at 12:58 PM, <[email protected]> wrote:
> Dear developpers,
>
> I wonder if some tool which would permit to recover
> an approximation of the density matrix from the available charge density
> is
> i) already available?
> ii) in preparation?
> iii) technically possible/stable/unique ?
> The idea is, to construct a density matrix in completely different basis,
> say by least squares fit, from a previous RHO, generated on
> a sufficiently dense mesh.
> This would permit to switch to different basis "on the fly",
> without the need to start from scratch, that may be annoying for
> big or delicate systems.
>
> Best regards
>
> Andrei Postnikov
>
>
