[CP2K:724] DFTB

[Bálint Aradi]

Hi,

  my name is Bálint Aradi and I'm one of the developers of the DFTB+
code (http://www.dftb-plus.info) and the one in charge for
distributing
SK-files on the web site http://www.dftb.org.
I'm not sure, if Juerg Hutters' was referring to our group in his
response, but since we're also using those "modified" SK-files,
I'd like to comment on that.

> parameter sets. We can read one type of data sets
> comming from the Dresden group. The parameters available are
> limited. If you want to use other data sets you have
> to fight with other (rather similar) formats and with
> copyright issues (yes, some groups don't allow you to
> use the parameters with other programs).

 DFTB+ (and other derivates/successors of the old Paderborn DFTB
implementation)
use a slightly different format from that of the Dresden group. This
is necessary
as Hubbard parameters (chemical hardnesses) are allowed to be
different
for different angular momenta. Furthermore, ususally splines are used
to describe the
repulsive part of the parameters, so that the appropriate fields for
the polynomial
repulsive are often empty in those SK-files.

  However, we are definitely interested in allowing people to use our
SK-files with
as many codes as possible. The license conditions do not restrict the
usage
in other codes at all! You're just not allowed to incorporate them
into some program and distribute
that program with them together. But you're free to download them and
use them with any
code capable to read them. Writing an appropriate extension for CP2K
should not
be very difficult, and I'd be definitely interested to provide help in
that.

  Best regards,

    Bálint Aradi