ATKError: Species not found

[kh]

Hi all,

I want to do a two-probe calculation with and si/sio2 system. I have used the vnl2.0 Tool to create the input file for the atk calculation. The atom coordinates has been done manually. For tunneling spectrum I inserted a phosphorus atom into the si electrode layer. After the self-consistent loop I get an error:
>>>>
Traceback (most recent call last):
  File "<string>", line 268, in ?
ATKError: # SpeciesContainer::attributes(Atom const & at) : species not found.
<<<<
I have tried to search this error in the manual, but I was not successful. And I do not know what the error should tell me, because the sc-loop is running. Has somebody the same error and could give me a hint?

Regards,
k.

[Anders Blom]

Can you share the script with us, it makes it a lot easier to troubleshoot.

[Anders Blom]

To clarify, this is an unusual error, which doesn't seem related to atom types (like if you just mistyped Phosphorus, which is easy enough to do!), as in that case the file wouldn't run but just return a Python error. This is a C++ error, and might instead indicate a corrupt NetCDF file or something similar.

If you have possibility, I strongly recommend you to upgrade to VNL and ATK 2008.10. You don't specify which ATK version you use, but I assume it's an older version.

[zh]

There is a hint. The surface layers in scattering region (or center region) are required to have the same characteristics (e.g, atomic structure) of electrodes because of the screening approximation to decouple the interaction between the electrode and the conductor of center region. Since  a phosphorus atom is inserted into the electrode layer,  a phosphorus atom should also be inserted correspondingly into the surface layer. 

[kh]

Hi,
that was the solution, thank you.

Regards