Simulating polymer

Dear LAMMPS users,
I want to simulate a system which contains water and polymer (poly acrylamide). My question is that can we start simulation with a box that contains both water and polymer or should we equilibrate polymer separately (so that it become polymer) and then mix it to the water?
Many thanks in advance

If possible, you want to start from an initial geometry that is as close to the final equilibrated state that you want to study. Your question sounds like you have little practical experience and thus it is highly advisable to find yourself and advisor/tutor/mentor that does have that experience and thus can guide you personally. A software specific forum on the web is not a good way to obtain that kind of advice.

I wrote a tutorial in which a polymer (a bit simpler than poly acrylamide may be) is merged with a reservoir or water. Both polymer and water were equilibrated separately. You can certainly use it as a source of inspiration.

Another clean way is to equilibrate the polymer alone, and then add water within the free space using the create_atom command with the overlap keyword.

A third way, that can eventually work too, is to put everything in a box and equilibrate the system.

1 Like

I dont know if this is of any value in your case, but there is this paper on a protocol to reach representative polymer samples (https://pubs.acs.org/doi/10.1021/ma200345v). The authors were dealing with a more rigid polymer though, so maybe it wouldnt be of any use in your case (for you to see).

Basically the paper talks about the challange of generating “real” (representative) amorphous structures since you dont have recipe on where to put the chains relative to one another, as it would be the case if you had a crystalline material. Also, polymers may have some slow dynamics, which can make it more difficult for an initial microstate generated randomly (or artificially) to converge to something “representative” by simply “waiting a long time”. So they develop an equilibration procedure that allows you to converge to a system that matches experimental results (structure, fits the broad range of densities, etc), so that it is validated in that sense. They talk a bit about in silico polymerization, but then you can skip that.
PS: but be mindful if you are planning to put everything in a box (i.e., water also) and equilibrate the system following that procedure because the equilibration has some agressive (T,p) conditions.

1 Like