Issues with geometry optimization of Iodine passivated Germanene Nanoribbon

[agoldsto]

Hello, I'm doing studies on edge passivation of Germanene Nanoribbons using a variety of elements. With Hydrogen Geometry optimization went smoothly, but I'm struggling to get a good result with Iodine as the passivating element. I'm using the same settings as I did for Hydrogen, that is:

- Rattle atoms prior to script generator
- In the New Calculator Block use: ATK-DFT, GGA-PBE Correlation, 75 Hartree density mesh cutoff, 1x21x21 k-points, Tight Tier 1 basis set (I have also tried double zeta polarized)
- In the geometry optimization block use: 0.01 Force Tolerance, 0.0005 Stress Tolerance, 200 steps, max step size of 0.5A, constrain cell in x direction

I have also tried adjusting the lattice parameters to increase size of unit cell prior to calculations and various force and stress tolerance values. I get different results, but all clearly incorrect. Any advice would be appreciated.

I've attached an image of the nanoribbon prior to rattling and optimization, and the post optimization result

[Anders Blom]

From a pure chemical argument, Iodine atoms will probably not like to sit like that at the end of the Ge bonds, so your result might actually be quite correct.

Is there any experimental evidence indicating the Iodine can passivate the edges of a Ge nanoribbon? Hydrogen atoms passivate the dangling bonds by "donating" an extra electron to create an electron pair, but Iodine sits at the other end of the periodic table, and would rather get an electron than give one away, so the chemistry of H and I will be very different.