Question Regarding Mean-Square Displacement (MSD)

Hi,

I'm having trouble understanding the MSD plot of a gas penetrant in a polymer matrix. I've used Forcite -> Dynamics -> Mean-Square Displacement on a trajectory (.xtd) file resulting from a 1000 ps simulation with NVE ensemble.

Multiple MSD plots from different models I have built result in a drop at the end of the Mean-Square Displacement Dynamics, as shown in the figure below. Since this has been a trend in quite a number of the plots (not all exhibit this behavior), I was wondering if someone could please help me understand why this is happening.

MSD plots.jpg

Is this because the NVE simulation is relatively short to produce a MSD plot that displays the normal diffusive regime, or is this trend due to the Mean-Square Displacement algorithm in Materials Studio? Or could it be because of the way I have built my models?

Any help would be greatly appreciated!

Best,

Matthew