How to fix gibbs ensemble

[Santoshph94]

Hi All,

I have a small doubt.

We can use fix nve for the microcanonical ensemble(NVE).
We can use fix nvt or fix nve and fix langevin for the canonical ensemble(NVT).

I want to run an ensemble where the chemical potential( μ), pressure(P), and temperature(T) are fixed.
How can I do this kind of ensemble simulation in the LAMMPS?

regards,
Santosh

[simongravelle]

If you are referring to the Gibbs ensemble, it was mentioned here a long time ago:

Gibbs Ensemble

Hi, I have seen in the archives that there may have been plans to implement Gibbs Ensemble MC into LAMMPS (http://lammps.sandia.gov/threads/msg43717.html) . Are there any immediate plans to do this? If not, I'm going to give it a go. Do you still recommend basing this on temper with atom swaps between boxes based on atom/swap or gcmc fixes? Thanks.

I have never seen it in LAMMPS, but I may be wrong.

Simon

[Santoshph94]

Thank you, Simon, for the quick response.

Yeah, I have seen that grand canonical Monte Carlo (GCMC) command in the LAMMPS.

I am doing MD. I am trying to make a condensate using the cubic box. Basically, what I want is to make a two-phase liquid and vapour phase coexist in the cubic box using a grand canonical ensemble.

Santosh

[simongravelle]

In that case, you should take a look at true Monte Carlo codes (GOMC, iRASPA, and many more). Although LAMMPS has some Monte Carlo moves implemented, it focuses more on molecular dynamics.

[Santoshph94]

Ok,
I will take a look at GOMC and iRASPA. I will also try LAMMPS GCMC.

thank you, Simon.

[stamoor]

We are currently working on implementing Gibbs Ensemble Monte Carlo (GEMC) in LAMMPS. We hope to have it completed in the next few months.

Basically, what I want is to make a two-phase liquid and vapour phase coexist in the cubic box using a grand canonical ensemble.

This sounds more like a two-phase MD simulation. GEMC has two simulations boxes for the two phases.

[Santoshph94]

Thank you, Stamoor.