Using "FIX one way" along with "FIX rigid"

Hi all

I was trying to study the influence of co2 on polymer coating. Iam treating the system as 2D periodic and i want to include long range electro static interactions through kspace_modify slab . In order to prevent the escape of co2 in +z direction, “iam using fix one way -z” on top portion of simulation box. since i have to use rigid co2 molecules, co2 is integrated by “fix rigid NVT”. but simulation is stopping after certain steps saying atom is missing(Bond atoms 823 825 missing on proc 0 at step 12796).
does this indicate both of these fixes cannot be used simultaneously?

What can be a better alternative for this .

Thank you

Whether fix oneway can work as a reflective boundary crucially depends on the width of the zone in which it is applied. So you may have to expand the box and with it the region in which the fix is applied for it to be effective.

An alternative would be a wall fix like fix wall/harmonic. Also here, though, you have to make sure that there is enough space for the wall forces to have an effect reversing the direction of the atoms/molecules without atoms leaving the box and choose the wall force parameters wisely.