On Computing the properties of mixture with Lammps

Dear all,

I am pretty new to molecular simulation. I just read some fundamental
knowledge on molecular simulation and the manual of lammps, while it
seems that it will take some time for me to get some clues.

My problem is the evaluation of the properties of the CO2 and water
mixtures (only single-phase considered at the moment), maybe Na+ and
Cl- ions later to handle salty water.

We need to get the change of the density, diffusion coefficients and
viscosity of the mixtures with the change of the concentration of the
CO2 in the mixture. The pressure and temperature are specified at the
moment.

Could somebody provides some suggestion on how to handle this kind of
computation with lammps? Maybe I can begin with the simplest one, the
density.
Is there any examples in lammps related to my problem closely? The
terms in molecular simulation are different with those in CFD and I
kind of got lost.

Thank you very much.

Best Regards,
Kai

Dear all,

I am pretty new to molecular simulation. I just read some fundamental
knowledge on molecular simulation and the manual of lammps, while it
seems that it will take some time for me to get some clues.

this is normal with all new skills.

My problem is the evaluation of the properties of the CO2 and water
mixtures (only single-phase considered at the moment), maybe Na+ and
Cl- ions later to handle salty water.

We need to get the change of the density, diffusion coefficients and
viscosity of the mixtures with the change of the concentration of the
CO2 in the mixture. The pressure and temperature are specified at the
moment.

Could somebody provides some suggestion on how to handle this kind of
computation with lammps? Maybe I can begin with the simplest one, the
density.

no. where you have to begin is to research the published literature.
this is simple enough a system that people must have made simulations
and experiments to compare them to.
more importantly, before you can start any classical MD simulations,
you need a parameterization that is describing the entities that you want
to simulate accurately enough and is transferable across a variety of
states. in your case, you thus need a parameter set for both, water
molecules and co2 and - most importantly - the interaction between
co2 and water. whether this parameterization is actually representing
the properties that you quote well enough (and don't forget solvation
shell geometry as represented by radial distribution functions, and
orientational relaxation times), is the typical measure of quality. to
that end, you will already find difficulties to find a water model that
is representing water well for *all* properties (pairwise additive water
potentials, don't "get" the density anomaly near the freezing point,
and are often overstructured or have the wrong density). there are
*lots* of review articles about water alone.

Is there any examples in lammps related to my problem closely? The
terms in molecular simulation are different with those in CFD and I
kind of got lost.

lammps doesn't ship any example for determination of density,
but all you'd have to do is to set up a system and equilibrate it
and then run for a long enough time in a properly set up NPT
ensemble and compute the averaged density of the equilibrated
part of the trajectory. *very* straightforward.

the major difference to CFD is that parameterization is much more
indirect and subtle. what you use as input in CFD is output in MD.
the understand the relation between those requires that you train
your self in statistical mechanics and statistical thermodynamics,
which is quite a challenging task and should not be taken lightly.
it is easy to run MD simulations at all, but it takes skill, practice
and knowledge to run *good* simulations and it can be even more
difficult to extract useful information from simulations. the rule of the
thumb is that the work *begins* with the end of a simulation, when
you try to make sense of it.

cheers,
    axel.

Dear Axel,

Thank you very much for your response.

I will try to train me more to make the problem more clear.

Regards,
Kai