RDF suddenly drops from 1

User discussions
1

GROMACS version:5.0.4
GROMACS modification: No

Could you please assist me in troubleshooting an issue with calculating the radial distribution function (RDF) for a system containing 34816 hydrogen atoms and a trajectory with 5412 frames and box length 8.4nm? I’m using the command ‘gmx_mpi rdf -f traj_comp.xtc -n atoms.ndx -o rdf-1.xvg’ on a node with 24 cores. However, I’m observing that the RDF looks unusual towards the end, dropping abruptly from 1 after reaching 2.7 nm. I’ve tested this with GROMACS version 4.5.5 and encountered the same problem. Any guidance on resolving this issue would be greatly appreciated.

h-h
h-h928×733 33.5 KB

2

I’m sorry that I am not sure about the specific reason without the information of simulation systems, but I guess the problem due to the maximum distance used in the RDF with incorrect pbc conditions. Maybe the description of gmx rdf would help you:

“The -rmax can be used to limit the computational cost if the RDF is not of interest up to the default (half of the box size with PBC, three times the box size without PBC).”

The g(r) converges to 1 after 1.5 nm in your figure, so I think the data previous of 2.7 nm is enough to describe this system.