[ase-users] [gpaw-users] Help calculating contact resistance

[jingzhe Chen]

Hi Alex,

          Normally the keywords nleadlayers, lead_atoms and mol_atoms
are seldom used,  code determines their values depending on the
pl_cells, pl_atoms input.  Using these keywords helps improve
efficiency, not the results.  For your system,  it helps if the tube is
 long like 2-3 nm.  I  will introduce these keywords in the wiki if you
need. Anyway, I have not seen the your error, can you show me
your structure and script?

        Best.
        Jingzhe


2011/8/25 Alex Eftimiades <alexeftimiades at gmail.com>

> Hello,
>
>        I am trying to calculate the contact resistance of two carbon
> nanotubes as follows:
>
>
> -------------------------------*
>
>        Tube 1
>
> -------------------------------*
>                        <----L---->
>                        *---------------------------------------------
>                                                Tube 2
>                        *---------------------------------------------
>
>
> As a function of "L."  The asterisks represent hydrogens at the end of
> the tubes. I have everything set up, but gpaw transport is giving me a
> strange error concerning "Value error: zero size array to ufunc.reduce
> without identity."
>
> Anyway, there are a lot of undocumented features in gpaw transport (I
> assume because the module itself is still under active development.)
> such as nleadlayers, lead_atoms, mol_atoms, etc. The method I used to
> set up the junction saves the following information:
>
> --Left and right lead atoms
> --L
> --the atoms that make up one unit cell of the each lead (and the unit
> cell itself)
> --the atoms that make up each lead
>
> Hopefully this type of setup is not particular to this study and
> others will find this thread useful.
> I not sure what variables should be set in the gpaw Transport
> calculator--I suspect mol_atoms is the scattering region, but I could
> not figure out how exactly gpaw Transport pieces together the whole
> setup (it seems to be able to guess parts of information.)
>
> I can extract more information if needed, but I need some help
> determining what variables I should set manually, and which should be
> left for gpaw to figure out automatically.
>
> Thank you,
> Alex
> _______________________________________________
> gpaw-users mailing list
> gpaw-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20110825/c7e22c4e/attachment.html>