[DFTB-Plus-User] R: Transport Calculation Issues with Poisson Box and Poisson gate

[ALESSANDRO PECCHIA]

Dear Pierre,

I think there might be a problem if you have a periodic system along y and ask for a planar gate, which is also placed along y.
Yet, it works with periodic systems such as a graphene or MoS2, where there is a dummy periodicity (sufficiently large) along y.
A way forward is to check your structure myself. There might be some unexpected 'problem' because the gates were not tested too extensively in the dftb+ code.
According to the error message it seems the Poisson solver does not converge or the operator turns out singular.
Indeed (from MUDPACK documentation):
ERROR
c     = -3  if the continuous elliptic pde is singular.  this means the
c          boundary conditions are periodic or pure derivative at all
c          boundaries and ce(x,y) = 0.0 for all x,y.  a solution is still
c          attempted but convergence may not occur due to ill-conditioning
c          of the linear system coming from the discretization.  the
c          ierror = -3 flag overrides the ierror=-4 and ierror=-5 nonfatal
c          flags.

It is not clear why it is singular since the gate is itself fixing a potential. Do you use localBC ?
You can send me an email to alessandro.pecchia at cnr.it

Regards,

Alessandro Pecchia

Da: DFTB-Plus-User <dftb-plus-user-bounces at mailman.zfn.uni-bremen.de> Per conto di Pierre.Cazade
Inviato: martedì 4 febbraio 2025 12:28
A: dftb-plus-user at mailman.zfn.uni-bremen.de
Oggetto: [DFTB-Plus-User] Transport Calculation Issues with Poisson Box and Poisson gate

Dear users,

I have issues when performing transport calculations on a periodic 2-lead system using a parallel virtual gate in the Poisson calculation. I systematically get the following error:

ERROR!
-> Poisson solver error n=-3

I have tried everything I can think of but to no avail. Is there a systematic approach (gate size or distance to the junction) to make these calculations work? Please note that my system is built so that the transport direction is along the z-axis, while the leads are periodic in the x- and y-directions.

I have a different issue when trying to do an SCC transport calculation on a non-periodic 3-lead system. The non-SCC works fine but the Poisson box fails to initialize when attempting a SCC calculation:

ERROR!
-> Failure during initialisation of the Poisson box

I have tried different box sizes from matching the exact size of the system to having a padding of 20 Angstrom on every side, again to no avail. Any suggestions?

Best regards,
Pierre
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20250204/2220343a/attachment-0001.htm>