GROMACS version: 2022
Dear GMX users,
I would like to perform a molecular dynamics simulation of a PEG polymer slab using the wall potential feature in GROMACS.
My idea is to start from an elongated simulation box in the z direction and then use repulsive walls on both sides of the box to gradually compress the system along z.
First, I equilibrated the system in the NVT ensemble using a box with dimensions 10 × 10 × 30 nm, without any wall potential.
In the next step, I added wall potentials in the .mdp file and planned to gradually reduce the box size in the z direction step by step (by 1 nm per step), for example: 10 × 10 × 29 nm, 10 × 10 × 28 nm, and so on.
However, in the first step, when I analyzed the trajectory, I observed that the polymer chains seem to move far from each other (to Z direction) without a clear, controlled compression of the system. I also calculated the radius of gyration (Rg), and it increases almost exponentially instead of showing a gradual response to compression.
Ideally, I would expect to see a controlled and gradual compression along the z direction driven by the wall potential and/or box reduction, but this behavior is not observed, and it seems to be reverse!
Could you please help me understand what might be causing this behavior and whether my setup or interpretation is correct?
I have also attached the simulation box scheme at the beginning and end of the NVT simulation in the first step.
; =========================================================
; WALL MODEL (for PEG slab compression in z direction)
; =========================================================
nwall = 2
wall-type = 9-3
wall-atomtype = WALL WALL
wall-density = 50 50
wall-r-linpot = 0.50
; PME slab correction (IMPORTANT for vacuum/slab systems)
ewald-geometry = 3dc
wall-ewald-zfac = 3
comm-mode = Linear
