Modelling evaporation-induced polymer aggregation

User discussions
1

GROMACS version: 2023.1
GROMACS modification: No
Here post your question

I want to simulate aggregation of polymer chains due to solvent evaporation using CHARMM36 FF. In several papers, this has been achieved in LAMMPS by sequentially removing solvent molecules which enter the vapor phase (using a distance-based criteria). These studies represented the solvent molecule as a bead.

I was wondering if using CHARMM36 FF in Gromacs will be a good choice. Apart from modifying the gro and top files after each solvent molecule removal, I assume I can directly move to the NVT production run.

Anyone having experience in such studies can please shed some light.

Thanks,
Raman

2

Hi,
I think the problem here is the ensemble, rather than the force field. How can you have an NVT ensemble if you are removing solvent molecules from the system?

3

Thanks for the reply @MichelePellegrino. The planned workflow is:
Step 1. Check solvent molecule(s) away from the solvent surface by x nm.
Step 2. The solvent molecule(s) at x nm away from solvent surface will be deleted from the gro and top file updated with remaining number of solvent molecules.
Step 3. An equilibration run followed by a production run for say, 100k steps.
Step 4. After 100k steps, return to steps 1 through 3.

I believe that a new NVT ensemble emerges after every 100k steps. Is this a reasonable way to achieve the desired result?

Regards,
Raman

4

I am not sure: evaporation is a rather fast and stochastic process, I am not sure if changing configuration files and re-running every time a molecule evaporates is a reasonable approach.
Some codes support muPT/muVT ensembles, probably you should look into that. Or you should find a way to simulate evaporation at NVT (i.e. without removing atoms/molecules).