# Reg: Calculating RDF of 5 different nanoparticles in a system.

Dear All,

I have a system which have 5 nanoparticles and water in the simulation domain. I took help from the LAMMPS forum and defined each nanoparticle as a chunk through chunk/atom command and then use compute com/chunk and fix ave/chunk to
write out the each nanoparticle’s centre of mass. So now I have coordinates of 5 COMs after N timesteps.

I am able to calculate the rdf of (H-H, H-O and O-O) through the following command:

compute HHrdf H2O rdf 1000 1 1
compute OOrdf H2O rdf 1000 2 2
compute HOrdf H2O rdf 1000 1 2

fix 4 H2O ave/time 10 100 1000 c_HHrdf[] file OOrdf.rdf mode vector
fix 5 H2O ave/time 10 100 1000 c_OOrdf[
] file OCrdf.rdf mode vector
fix 6 H2O ave/time 10 100 1000 c_HOrdf[*] file CCrdf.rdf mode vector

But I am not able to calculate the rdf of the above 5 COMs of the nanoparticle as they are not grouped as above. Please suggest me how can I perform the rdf calculation for nanoparticles within the system.

The reason I am interested in calculating the rdf of nanoparticles is to see how they aggregate when I increase the number of nanoparticles in the nanofluid system (Nanoparticles + water).

The reason I am interested in calculating the rdf of nanoparticles is to see how they aggregate when I increase the number of nanoparticles in the nanofluid system (Nanoparticles + water).

please note, that a g® for only a small number of particles has limited meaning.

the simplest way to proceed would be to just output a trajectory of the center of mass of your nanoparticles and then compute the g® with an external tool.
with so few particles, you don’t have to worry about performance, so you can compute it in a script language like python or matlab. there are - literally - 100s of examples of how to compute a g® posted somewhere on the web. …and there are plenty of software packages for MD analysis, that can do it. all you have to do is to make those packages believe, that you are feeding it a real trajectory of particles, instead of COM positions. hell, this might even work with LAMMPS using the rerun feature, if you “massage” the COM trajectory output properly.

axel.