Transmission of graphene ribbon is a pulse instead of spike around Fermi level

[Wanzhi Qiu]

Dear ATK users,

I have a issue in using DFT to calculate transmission T(E) of a graphene ribbon. The ribbon is zigzag of ~3nm width, a hole of diameter ~1 nm is created in its center by removing some atoms. Unexpectedly the calculated T(E) around Fermi level looks like a pulse instead of a spike. The calculate energy range is (-2 2) eV with sampling period 0.02 eV. The T(E) has two peaks at two adjacent energy points; having values 2.01 and 2.00 @ E=-0.02 eV and E=0.0 eV respectively. All the default parameters are used in the calculation, including k-point sampling (1, 1, 100) and mesh cutoff 75. The ATK version is 12.8.1. Any ideas on solving this issue? Thanks..

[Anders Blom]

What is the difference between pulse and spike in this context? In general I don't think there is any numerical issue, but without more information (like at least a picture of the system and results) it's impossible to help.

[Anders Blom]

I don't see any reason to think T(E=-0.02) should be zero. The results look reasonable.

[nori]

I guess that the band structure at E = 0 eV is not completely flat but a bit distorted numerically.
That is the reason why T(E=-0.02) = 2.01.

But I think your calculation is OK because this artifact brings no influence.

[Wanzhi Qiu]

Hi Anders and Nori, I take your advices, much appreciated! Regards, Wanzhi.