Dear Moon,
as is customary the coordinates in the output are not folded back into your cell in order to be able
to easily compute transport properties such as the translational diffusion constant via mean squared
displacements. Hence, even though internally periodic boundaries are properly taken into account
you may have the sensation that the atoms are spreading out when visualizing them.
If that is case everything everything is fine, but you can fold the coordinates back into the cell when
visualizing them.
Best,
Thomas Kühne
Thank you for your comments.The displacement of the atoms is indeed larger and larger, but they don't move to replicas of the box.
I am confused with this phenomenon.