[DFTB-Plus-User] Problem in Solving (100) silicon electrodes using DFTB+

[Gabriele Penazzi]

On 10/11/2013 06:23 AM, n.salami at srbiau.ac.ir wrote:
> Dear users,
> Good time,
>
> I have built an xyz file of a model, as described below, to calculate the
> transport properties of an SiNW in 100 direction ,where the electrodes are
> SiNW in 100 direction.
> I built the model by cutting silicon bulk such as I had 10 unitcells in x
> direction and 3 unitcells in y and z direction by vmd software
>
> http://www.ks.uiuc.edu/Research/vmd/
>
>
> and I have generated the xyz file with the precision of 10e-2 Angstroms.
> and then I have passivated by hydrogen.
> I got 8 layers of Si, 2 unitcells, for  the source and drain
> contacts,seprately , and I got 6 unitcells for the device.
> I have built the gen format of geometry by xyz2gen script.
> Then, I ran the file without minimization, since I calculated the
> coordinates precisely.
> However, complying all the model generation rules for this software, when
> I solved the electrodes by DFTB+, I faced the error below:
>
> "Contact 1 does not consist of two rigidly shifted layers.”

Hi Nadia,

Check for the parameter ShiftAccuracy in the manual,
which tells how "rigidly" the layers need to be shifted.

The default value is 1e-8 a.u., very low, to try to force the user to
build a structure exact up to numerical precision
(i.e. by shifting the layers by himself or blocking 2 layers during
relaxation). In this contest 1e-2A of displacement is not negligible.

Instead of building the whole structure and relax it, you can try to
build a single layer and repeat it, this will force numerical precision.

Or you can define a higher ShiftAccuracy value, but I don't suggest you
to do this, you could (probably will) get some fictitious features in
the transmission.

Best,
Gabriele



>
>
> I appreciate your consideration in anticipation.
>
> Regards,
> Nadia Salami
>
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at dftb-plus.info
> http://www.dftb-plus.info/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
fax: +49 (0) 421 218 62770