May I know whether Amorphous cell packing is a random process? I noticed that sometimes when I submitted input files for packing followed by geometry optimization, the job finishes with the warning message “optimization with current method stopped”. But when I submit the same input files again there is a chance that optimization is achieved without any error.

I'm modeling a CNT embedded in PE matrix and I'm employing periodic boundary condition. Is there any way I can place the CNT precisely at the center of the simulations cell before I start packing PE chains in to the cell? For the moment, I manually drag the CNT in to approximately the center of the simulation cell and I doubt whether that could affect the precision of the results.

Thanks in advance