Polymer becomes immobile when center of mass constained

LAMMPS
1

Hello,

I am trying to simulate the interface between an amorphous polymer melt and a rigid graphene sheet but I am finding the polymer seems to “freeze” and become completely immobile during the NVT run, despite the thermo output data showing normal temperature and kinetic energy throughout. In my simulation I am using GAFF parameters for both the polymer and graphene, Nose-Hoover thermostat and “fix recenter” command to constrain the center of mass of the polymer.

For debugging, I have created a small test system containing a single polymer chain and a graphene sheet. I have tried rerunning using different thermostats and found the same problem occurring with Berendsen and velocity rescale but Langevin seems to work fine. I also noticed that the polymer doesn’t “freeze” when I don’t include the fix recenter command, however I need to prevent system drift in order to study diffusion properties.

Any insight into why this is happening would be much appreciated. I have attached the data file, force field files and input script.

Many thanks,
Nisha

(LAMMPS version: 29Aug2024)

polymer_graphene.data (224.2 KB)
run_nvt.in (1.2 KB)
system.in.init (333 Bytes)
system.in.settings (2.3 KB)

2

Fix recenter does not do that. It merely shifts the coordinates so that the center of the fix group remains in place. So it is not a constraint, but using a different point of reference. In fact, you cannot have a “position constraint”, all that you can do is a position restraint and that can be done for a group with fix spring or for individual atoms with fix spring/self.

This hints at being a victim of the so-called “flying ice cube syndrome” (search the forum for more details on that). Fix recenter does not solve that, but only hide it.