Non-self-consistent and Equivalent bulk

[merlin021]

Hello, Sir,

I feel confused about descriptions which seem to contradict with each other as follows.

1. It says "It is not allowed to use NonSelfConsistent for EquivalentBulk."
(https://docs.quantumwise.com/manuals/Types/EquivalentBulk/EquivalentBulk.html)
2. It also writes "Setup a non-self-consistent calculation with an equivalent bulk initial density."
(https://docs.quantumwise.com/manuals/Types/DeviceAlgorithmParameters/DeviceAlgorithmParameters.html)
3. There seems a calculation done with "non-self-consistent with equivalent bulk initial density".
(http://quantumwise.com/documents/tutorials/ATK-11.8/BasicGrapheneTutorial/index.html/chap.zigzag_transport.html#sect2.zigzag.equiv_bulk)

Would you please elaborate a bit about this ?
Thank you very much.

[Ulrik G. Vej-Hansen]

You are right, there seems to be an inconsistency there - I will look into it.

However, point 3. is a tutorial for a very old version of ATK, so that is no longer relevant. Tutorials not on docs.quantumwise.com should be considered obsolete unless you use a quite old version of VNL/ATK.

[merlin021]

You are right, there seems to be an inconsistency there - I will look into it.

However, point 3. is a tutorial for a very old version of ATK, so that is no longer relevant. Tutorials not on docs.quantumwise.com should be considered obsolete unless you use a quite old version of VNL/ATK.

Hi, sir, thanks for reply and investigation.
yes, that's an old version, and I'm now using version 2016.4.

[Ulrik G. Vej-Hansen]

I have looked into it, and the current writing is correct.

Point 1. refers to the fact that you can define the IterationControlParameters for the Equivalent Bulk calculation itself, and those are not allowed to include NonSelfConsistent, which means that the EquivalentBulk calculation itself must be self-consistent - otherwise it would essentially just be the same as NeutralAtom.

However, point 2. refers to the fact that you can do a non-selfconsistent calculation, using the result of the EquivalentBulk calculation as the starting point.

Did that clear things up for you?

[merlin021]

I have looked into it, and the current writing is correct.

Point 1. refers to the fact that you can define the IterationControlParameters for the Equivalent Bulk calculation itself, and those are not allowed to include NonSelfConsistent, which means that the EquivalentBulk calculation itself must be self-consistent - otherwise it would essentially just be the same as NeutralAtom.

However, point 2. refers to the fact that you can do a non-selfconsistent calculation, using the result of the EquivalentBulk calculation as the starting point.

Did that clear things up for you?

Appreciated, I am convinced by that.

Can I ask one additional question that, does it matter if the Fermi level of the calculated electron transmission spectrum is NOT in the middle (at 0 eV energy) of of the conduction band minimum and valence band maximum, provided that I've set energy zero parameter as "Average Fermi level", with which the Fermi level is supposed to be in the middle. And what do you think would cause this?

Thank you very much.

[Ulrik G. Vej-Hansen]

If the Fermi level is not in the middle of the band gap, that is usually because the number of k-points is too low. Please make sure k-points etc. are converged - it will improve the quality of your results.

[merlin021]

If the Fermi level is not in the middle of the band gap, that is usually because the number of k-points is too low. Please make sure k-points etc. are converged - it will improve the quality of your results.

I'd check that, thanks very much.