[DFTB-Plus-User] Why the Borg-0-1 parameter set gives a negative repulsive potential value?

[Peter Yen]

Dear Mr. Bálint:
Thank you for your reply. As a matter of fact,  using this parameter set, i
can't  predict the structure of double ring structure for B_20. Double ring
 is commonly accepted as the global minimum structure for B_20. This
structure is predicted by other groups using B3LYP+ 6-31G(d) ,  which is
also used for fitting the DFTB parameter set  in Borg. So i assume that
 Borg can predict the double ring structure as well, at least not too far
from global minimum. However , the energy of double ring structure
is unreasonably high(total energy 20 ev higher than the global ).
Therefore i try to inspect what is the problem inside and i discovered that
the repulsive energy is a negative value. I am not sure if this is a bug in
Borg set but this is something i have tested so far. And I will be grateful
if you can provide me with any further suggestions.



 Best Regards
                                                                         Yen



2014-03-19 17:10 GMT+08:00 Bálint Aradi <balint.aradi at bccms.uni-bremen.de>:

> Dear Yen,
>
> > The reference "SCC-DFTB parameterization for boron and boranes. J. Chem.
> > Theory Comput. 8, p. 1153" states that the parameter set is applicable to
> > finite and perodic systems. I assume that this means it can be used for
> > boron clusters. Also in dftb website they claim that this parameter set
> is
> > fitted to molecular system so i think it should work even better for
> boron
> > cluster. Anyhow, it shouldn't generate a negative value for "repulsive "
> > energy. Forget to mention, my system is B_20 and the input structure is
> > obtained from a global optimization program.
>
> In ideal cases, the repulsive should be positive indeed. However, if you
> look on the repulsive of Bohr as stored in the B-B SK-file (see attached
> figure) you see, that it is negative in a wide region (Energies are
> given in Ha, lengths in Bohr), so no wonder that you obtain negative
> repulsive energies.
>
> Now, whether a negative "repulsive" potential is always an evil thing or
> not, is a different question. I am more for the pragmatic approach: The
> repulsive potential just contains all terms which are not explicitely
> calculated in the DFTB approach + all corrections compensating for the
> errors we did in the explicitely calculated (but strongly approximated)
> terms.
>
> So, the advice is always the same: just compare the results for small
> systems agains ab initio results. If they are fine and you get
> qualitatively good results, scale up to bigger systems.
>
>   Best regards,
>
>   Bálint
>
>
> --
> Dr. Bálint Aradi
> Bremen Center for Computational Materials Science, University of Bremen
> http://www.bccms.uni-bremen.de/cms/people/b_aradi/
>
>
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at dftb-plus.info
> http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user
>
>


-- 
Research Assistant,Physics
Department,NCU,Taiwan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20140320/3abf1968/attachment.htm>