How to remove artefact?

[Leon Wang]

Hi, Bachir,

I refined some structures. There is a artifact in the G(r). I discussed with Peter and updated rmin and O-O distance. However, the artifact is still existing shown below. Attached is my run.py.  Any suggestion would be highly appreciated.

Appreciatively,

Leon

[Bachir Aoun]

Hello Leon,

i can't tell why you have this peak but if it's there that means you have some important O-O correlation at this 

distance.

I am not sure what you implemented in those functions so I can't help with that part

from moves import atoms, tunnels, remove_heavy, swap_frame_tunnel

Have you considered swapping and removing oxygen ?if you can't get rid of this peak there is maybe a reason for that which is more oxygen in your system than needed.

bachir

[Leon Wang]

Hi, Bachir,

Thanks for your reply. I double checked the atomic number of each element. No change occurs in the initial and final structures. I also checked the stoichiometric values and ratio. There are not more oxygen in the system.

I discussed with Peter.

"tunnels moves only tunnel species; remove_heavy removes Cs, Ba from the tunnel; swap_frame_tunnel tests  swapping tunnel and framework species."

Apparently, there are some other reasons causing this artifact. If there were no error in the structure, do you think that it is the possible bug in fullRMC?

In my three refinements, only 1-07-1-mono does not have this artifact. The rest of two have this artifact. If you need, I can provide you the data files for your test or debug.

Best,

Leon

[Bachir Aoun]

Leon,

there is always the possibility of a bug in fullrmc but this one doesn't seem like it. 

I have hard time doubting a bug in the pair distance calculation. This would have cause problems a long time ago. You can double check it by yourself actually outside of fullrmc. Export your structure and see whether you have an intense peak between oxygens.

In my opinion, your problem is either in the stoichiometry of your system or the boundary conditions. This can also happen when you create a supercell and you're not careful about the atoms at the boundaries.

Regardless, if you want you can share your data but i can't promise working on it as fast as i used to. i am very busy with deadlines now

thanks