[lammps-users] Bond atoms missing!

To avoid overlaps I use the soft potential as in the script. I could get the poiseuille flow with this same input file when the melt has less chains but as soon as the I increase the
number of chains it gives the error. I do not understand why is this. Played around with the
parameters of soft potential and tried energy minimization, but no success yet.

The next question is when I get a poiseuille flow with few chains, initially the chains are spread out in the channel. When the pressure is applied along the x-direction the chains start to come closer and finally form a clump leaving vast amount of channel empty. We want to fill the channel with uniform density. How can this be done? Any suggestions would be of great help.


If you don't un-overlap long enough, and then start MD with a hard
potential (e.g. LJ), then you will move atoms far away in 1 or 2 timesteps
and get the lost atoms error.