NEMD Simulation of Planar Elongational Flows

LAMMPS LAMMPS Mailing List Mirror
1

Hi,

I am trying to perform a nonequilibrium molecular dynamics (NEMD) simulation of a polymer melt under planar elongational flow (PEF). To do this I need to use “fix deform” to change the simulation box shape. Here is an example of the fix command that I use:

fix 1 all deform 1 x trate 0.001 y trate -0.001 remap v

Obviously the simulation box shrinks in y-direction with time, and after a limited time it becomes so small that I have to stop the simulation. This time is usually too short comparing to the relaxation time of the polymer melt so that I am not even able to equilibrate this system.

One of the possible solutions for this problem is applying the Kraynik-Rainelt boundary conditions which are both spatially and temporally periodic. These periodic boundary conditions guarantee theoretically infinite simulation times and have been used frequently in NEMD simulations of planar elongational flows.

I was wondering if Kraynik-Rainelt boundary conditions are implemented in LAMMPS? If not, is there any alternative method in LAMMPS that provides an effectively infinite simulation time for PEF simulations?

Thanks,

Hadi

2

I was wondering if Kraynik-Rainelt boundary conditions are implemented in LAMMPS? If not, is there any alternative > method in LAMMPS that provides an effectively infinite simulation time for PEF simulations?

KR is not implemented (yet) in LAMMPS and there is no alternative

that lets you run for infinite. We had preliminary discussions
a year ago with a group interested in working on this, but nothing

has resulted yet. It is not a simple algorithm to implement in LAMMPS
as it involves both changes to the box shape and rotations of the atoms.

Aidan can comment on the idea/status. Maybe
you want to help?

Steve