[DFTB-Plus-User] How to generate s-k files?

[Gabriele Penazzi]

On 05/11/2015 10:09 PM, Michael Foster wrote:
> I have a related question involving s-k files. Ideally, I would like
> to model a system containing Zn, C, H, N,  and O on a TiO2 surface.
> Slater-Koster files exist for all interactions except for Zn-Ti;
> however, in my system the Zn and Ti atoms are well separated.
> Therefore, it would seem reasonable to neglect these interactions; is
> there a way to do this in DFTB+? Can I create a generic s-k file with
> zeros in it or something like that and name it Zn-Ti.skf?

Hi Michael,

in line of principle I think you can create a dummy sk file. Anyway, are
you mixing different sets? If yes, this is not a safe practice:
different sets may have different energy alignment, use different basis
set for a give specie etc.

>
> Another related question. Suppose you would like to model a system
> containing many porphyrins with different metals within the porphyrin
> ring for which no s-k files exist. Would it be relatively straight
> forward to create s-k files for a specific system? I know great care
> and considerable time is put into creating the mio and 3ob parameters
> so that they work for a diversity of systems. But if I only want to
> perform calculations on a specific system, such a porphyrin with Pd in
> it, one would think that you could extract the relevant information
> from a simple DFT calculation. Any thoughts? 

Take my words with a grain of salts, I do not develop parameters. But I
do run charge transport calculations, and in that case we do what you
mention often: we tune parameters on a more specific problem, for
example to get better fit on some specific symmetry points of the band
structure or better transmission decay etc.

Of course accuracy, portability and complexity are not linearly
independent variables, and if you try to tune parameters too tight on a
specific test set, you have to be careful when you add afterwards
additional degrees of freedom in your study.

DFTB, having a small number of fitting parameters, usually may ensure a
better portability but you cannot escape the fact that the smallest the
target and the tighter the fitting, the lesser the portability
(normally). It is good also in that case to have some kind of validation
system. Otherwise the distance between DFTB and fully empirical becomes
somewhat short.

Best
Gabriele


>
> Thanks in advance for a comments    
>
> On Mon, May 11, 2015 at 6:28 AM, 李捷 <lijielile at gmail.com
> <mailto:lijielile at gmail.com>> wrote:
>
>     Hello, Professor Aradi,
>     Thank you very much for your reply and information. I definitely
>     would like to try those files you mentioned but I can't find them
>     from the link you sent. I appreciate it a lot if you can let me
>     know how can I get them? We will definitely acknowledge your
>     contribution in any resulting publications. 
>
>     Thanks again for your kind help.
>
>     Best
>
>     Jie
>
>     On Mon, May 11, 2015 at 5:40 AM, Bálint Aradi <aradi at uni-bremen.de
>     <mailto:aradi at uni-bremen.de>> wrote:
>
>
>
>         On 05/08/2015 03:43 PM, MF wrote:
>         > I have never used it but Materials Studio says it can
>         generate s-k files if you have DFTB+ and Dmol. Although, it
>         will cost you $$$. Maybe the developers can comment more on this.
>
>         That's true, Material Studio contains a nice GUI over the
>         generation
>         tools, making it simpler for users to generate SK-files. The
>         tool was
>         written by the guys at Accelrys/Biovia, and therefore, it is
>         available
>         only as part of Materials Studio (which indeed is not free of
>         charge).
>
>         Once we have a similar, well tested, properly working wrapper
>         (although
>         non-graphical), it could be released to the public for free of
>         charge.
>
>           Cheers,
>
>            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 mailman.zfn.uni-bremen.de
>         <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>
>         https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
>
>
>
>     _______________________________________________
>     DFTB-Plus-User mailing list
>     DFTB-Plus-User at mailman.zfn.uni-bremen.de
>     <mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>
>     https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
>
>
>
>
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at mailman.zfn.uni-bremen.de
> https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user

-- 
--
Dr. Gabriele Penazzi
BCCMS - University of Bremen

http://www.bccms.uni-bremen.de/
http://sites.google.com/site/gabrielepenazzi/
phone: +49 (0) 421 218 62337
mobile: +49 (0) 151 19650383

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20150513/6746bbb1/attachment.html>