I Started to study the penetration of carbon atoms into iron oxide. but I encountered a problem
My simulation box contains an iron oxide cylinder shell surrounded by carbon atoms. After a few steps, the near-surface atoms penetrate into the oxide layer and end up creating a gap between the rest of the carbon particles (farther away) and the shell. Therefore, those particles do not participate in the diffusion.
Is there a way for carbon atoms that are far from the shell to approach and participate in the diffusion?
It is difficult to discuss such items without know more details.
What you describe and what is shown in the images is rather unusual.
What kind of state are the carbon atoms supposed to be in? solid, liquid, or gas?
and at what temperature?
For a gas the density seems far too high.
Do the other carbon atoms move?
The fact that they get attracted by the oxide seems to be due to the potential between the oxide and the carbon atoms, but they potential seems to have a cutoff and thus only atoms within the cutoff are pulled toward the cylinder.
But that would also imply that the carbon atoms individual don’t move around on their own.
There can be multiple reasons for this behavior. Most likely are a bad simulation setup and unsuitable potential parameters.
Dear Doctor, In fact, I want to investigate the penetration of carbon atoms into an iron oxide nanowire (a type of catalyst). The carbon atoms are soluble inside a solvent, which I have discarded and placed only the carbon atoms(solid state) inside the simulation box.
I have used reaxff potential and because I did not know how to generate the electric current to move the carbon atoms to the layer, I set the system to 500 Kelvin.
After a few steps, some of the atoms move and the rest remain constant.
I have used :
neighbor 2.5 bin
neigh_modify every 5 delay 0 check no
fix Diff all nvt temp 500.0 500.0 40.0
I do not know what to do
Could you guide me, and if I have to do an electric current to move atoms, how should I do it?
Sorry, I cannot help you with this. This seems much more a problem of the “science” of how you need to plan and set up your simulation and what choices you have to make. Instead that is something you need to discuss with your adviser/supervisor or equivalent. Multiple of the choices you describe sound questionable to me, but I don’t have enough experience in simulating this kind of system to make any suggestions with confidence.