Questions in GCMC

LAMMPS LAMMPS Mailing List Mirror
1

Dear lammps users:

I am new to Lammps, and I want to do some simulations with the command “fix gcmc”. My simulation is about the adsorption of polyelectrolyte in sea water, i.e. there are polymer chain, water and NaCl in the simulation box. I want to know how many water and Na+ Cl+ can stay within the memberane of polymer chains. Currenly I know GCMC is the right method to be used, but I am not very sure about whether “fix gcmc” in lammps is the right software. My confusions are as follows:

Q1: I want to exhcange water, Na+ and Cl- with the outer reservoir, is this allowed? In the manul it is said that “use of multiple fix gcmc commands in the same input script can be problematic if using a template molecule”. So, what is the solution to exchange all the particles, i.e. water, Na+, Cl-?

Q2: It is said that GCMC can be done with “fix nvt” and “fix gcmc”. The former, “fix nvt” is with constant n,v, and T; while “fix gcmc” is only with constant mu and T, so what happens to the volume of system when run “fix gcmc”?

Q3: Molecules or atoms of exchanged type can be inseted or deleted, displaced, rotated, but all these treatment can be done on other molecules or atoms except insertion. That means, Molecules of not exhanged type can be deleted but not inserted? Am I understanding correct? Why the delete is not forbidden to molecules not of exchange type?

I hope my questions did not bother you. Thanks in advance!

Best,

Yongbiao

2

I am by no means an expert in the topic but will comment a bit to point out to some facts.
First, I suggest you look in the lit to find how people tackle similar problems from a methodological point of view. I’m sure this type of simulations are run often by people (the bio guys) doing solvated protein simulations, etc. My first concern is that if your polymer chain is neutral you automatically need to have equal numbers of Na and Cl ions thus you cannot attempt to untangle the insertion/deletion of anions from cations. Furthermore, I would expect people to know in advance the concentration (or at least the range of concentrations) of the ions in solution. This will make the number of Na/Cl ions fixed during the runs. With the two previous conditions holding, only the water (its chemical potential) would remain as the entity to be played with during the GCMC runs (from a lammps point of view). But which GCMC approach works best is another question. It may turn out you will need to use CBMC techniques as well to deal with your polymer dofs when searching for the equilibrium configs. To summarize, understand first very well the technical procedure to be followed in order to properly design a strategy as to which software(s) is(are) to be employed and how to set your systems for successful computations. Apologies for not handling down the exact recipe but never worked on this kind of simulations.
Carlos

3

Dear lammps users:

I am new to Lammps, and I want to do some simulations with the command
"fix gcmc". My simulation is about the adsorption of polyelectrolyte in sea
water, i.e. there are polymer chain, water and NaCl in the simulation box.
I want to know how many water and Na+ Cl+ can stay within the memberane of
polymer chains. Currenly I know GCMC is the right method to be used, but I
am not very sure about whether "fix gcmc" in lammps is the right software.
My confusions are as follows:

Q1: I want to exhcange water, Na+ and Cl- with the outer reservoir, is
this allowed? In the manul it is said that "use of multiple fix gcmc
commands in the same input script can be problematic if using a template
molecule". So, what is the solution to exchange all the particles, i.e.
water, Na+, Cl-?

​it is in general always a good idea to try to reduce degrees of freedom,
and i think that for your study having three independent exchanges is
making your problem needlessly complex. also, it is not quite clear, why
you need MD. if all you care about is the "loading" of your
polyelectrolyte, a pure MC simulation should do the trick just as well. if
you care about the concentration gradient between the (loaded?) polymer and
the seawater, then you don't need GCMC at all, and if you care about
transport, then i doubt that MD is covering time scales properly, so you
may have to resolve to kinetic MC or similar.​

Q2: It is said that GCMC can be done with "fix nvt" and "fix gcmc". The
former, "fix nvt" is with constant n,v, and T; while "fix gcmc" is only
with constant mu and T, so what happens to the volume of system when run
"fix gcmc"?

​this is a common ​misunderstanding. fix nvt doesn't *enforce* constant
"n", "V" or "T", it does *adjust* T (in contrast to fix nve, which doesn't
adjust anything), but you have constant T only if there is nothing else
manipulating the system in a way that may affect the kinetic energy. in
addition, fix nvt simply does *not* touch "n" or "V". similar for fix gcmc,
it *only* adjusts "n". so what you get as a result is the ensemble that
corresponds to the combined manipulations.

Q3: Molecules or atoms of exchanged type can be inseted or deleted,
displaced, rotated, but all these treatment can be done on other molecules
or atoms except insertion. That means, Molecules of not exhanged type can
be deleted but not inserted? Am I understanding correct? Why the delete is
not forbidden to molecules not of exchange type?

​it is not really clear to me what you are asking. the wording is
confusing. please reformulate your question. what fix gcmc does is what is
described in the documentation, nothing else, nothing less. there are other
fixes that can change the system (e.g. add/remove/swap​ bonds), but each of
these are usually independent and do not communicate any information about
applied changes. the latter is also an issue with using multiple instances
for fix gcmc, hence it should be attempted to reformulate a simulation in
such a way, that only one fix is needed to do MC-style changes.

​axel.​

4

Regarding Q3, if you are going to quote from the manual, you should
not attempt to paraphrase. Instead just provide the quote (which I
found by Bayesian analysis):

"This fix cannot be used to perform MC insertions of gas atoms or
molecules other than the exchanged type, but MC deletions,
translations, and rotations can be performed on any atom/molecule in
the fix group."

The answer to your question "Molecules of not exchanged type can be
deleted but not inserted?" is yes. In your example, the polymer
atoms should not be included in the fix group.

Exchanging Na+ and Cl- ions in neutral combinations is beyond the
scope of the LAMMPS gcmc fix. For that, you should consider using a
dedicated Monte Carlo code. If you are willing to limit your
simulation to exchange only water molecules, then fix gcmc should work
just fine.

Aidan