Calculating the spin polarization of a Zigzag GNR with DFTB

[Yueyang]

Hi,
I have tried to calculate the bandstructure of a ZGNR with the DFTB, the spin is set to be polarized,
but the calculated banstructure turns to be degenerate no matter  the ZGNR is ferromagnetic or antiferromagnetic,
Why?

[zh]

There is one trick for the spin polarization calculation of ZGNR, that is, only the edge carbon  atoms are required to specify the initial magnetic moments.
Have you check the spin distributions in your ferromagnetic and antiferromagnetic configuration calculations? How does they look like?
I guess that the setup in your ferromagnetic configuration may be not correct and the calculation may be not converged to the desired ferromagnetic state.

[Yueyang]

Hi Zh,
Thank you for your reply.
I did set only the edge carbon  atoms magnetic, but I'm not sure whether they converged as expected.
Below is the py file, please take a look at it.
Thanks again!

[zh]

Please refer to my calculation setup for the ferromagnetic state of ZGNR.

[zh]

The nc file.

[Yueyang]

The py file is much more complex than I expected, and I don't know how to generate it with the VNL.
Although the degeneration of the band is eliminated, the bandstructure is still quite different with the one calculated by DFT.
However, I just found that the DFTB works well on ZGNRs that passived by hydrogen,
I wonder wheter passivation is necessary for considering Spin with DFTB?

[Anders Blom]

It's actually not at all difficult, the only relevant line is the "initial spin" part and this can easily be set up in the script generator.