Potential of mean force can be calculated from radial distribution function (g(r)), from the equation: w = -kB*T*ln[g(r)]. RDF or g(r) can be calculated either from lammps "compute_rdf" command or directly from trajectory analysis.

Potential of mean force can be calculated from radial distribution function (g(r)), from the equation: w = -kB*T*ln[g(r)]. RDF or g(r) can be calculated either from lammps "compute_rdf" command or directly from trajectory analysis.

Free Energy of a given N particle system can have a dependence on the 6N degrees of freedom. Anylicatically or computationally the free energy surface as a function of these 6N degrees of freedom if not impossible is very hard to get. When studying some system computationally one may be interested to get the free energy as a function of some relevant degrees of freedom or reaction coordinate. The free energy surface along the chosen reaction coordinate is introduced to as the potential of mean force.

There are several advance sampling techniques (such as Thermodynamic integration, Umbrella sampling, Replica Exchange Molecular Dynamics, Steered Molecular Dynamics and the Jarzynski Relationship etc. ) to calculate the PMF.

You can implement any method in LAMMPS depending on your system and how much stat mech you know.

Dear Naskar

Thanks so much.

But I'm new to LAMMPS and don't know how to carry out those sampling methods.

Would you please introduce me some illustrative references or websites?

Thanks in advances

Hey Amir,

You can find a nice description of the software and sufficient tutorial to get started in the official website of the Lammps : http://lammps.sandia.gov/doc/Manual.html

Dear Kanka Ghosh ,

Can you please give any reference to the formula that you gave to compute PMF using RDF?

Thank you.