MSD

LAMMPS LAMMPS Mailing List Mirror
1

hi everybody
how MSD show that the system get to equilibrium?
I plot msd vs. time and it become a line with slope and arise by time
, but have a bit fluctuation , Is that mean system get to equilibrium
or it should be linear with zero slope?
my system is ionic liquid
thanks

2

hi everybody
how MSD show that the system get to equilibrium?
I plot msd vs. time and it become a line with slope and arise by time
, but have a bit fluctuation , Is that mean system get to equilibrium
or it should be linear with zero slope?
my system is ionic liquid
thanks

This is not a LAMMPS question.
And attempting to post a definition of what "equilibrium" means on the
LAMMPS mailing list is probably a bad idea.

Knowing when a system has equilibrated requires some insight about the system.

I am not sure why you want to use MSD to check for equilibrium.
The graph of MSD vs time will only have zero slope if the molecules
are not moving. If you are simulating a system in the liquid phase,
the slope should not be zero.

For a liquid of (approximately) spherically symmetric molecules, it
might be enough to wait for the volume to equilibrate. When
simulating non-spherical molecules, or flexible molecules like
polymers, it can be quite difficult.

Is the volume of the system continuing to increase? (or decrease?)

Andrew

3

My guess is that you meant when does the MSD plot converge to the diffusive regime. Still not a LAMMPS question. What I like to do is to plot the block averaged MSD vs time in a log-log plot to identify the diffusive region by eye.

-s-