Please post to the mail list, not to me.
Compute rdf and msd know nothing about
whether boundaries are periodic or not.
They work either way. Of course
they will give different answers, since
atoms aren’t diffusing freely in z, nor are
their neighbors of atoms in z to long distances
for the RDF.
Steve
Please post to the mail list, not to me.
Compute rdf and msd know nothing about
whether boundaries are periodic or not.
They work either way. Of course
they will give different answers, since
atoms aren't diffusing freely in z, nor are
their neighbors of atoms in z to long distances
for the RDF.
more importantly, for a homogeneous and fully periodic system, RDF and
MSD do not depend on the position of the respective particle of interest;
for a slab or confined, i.e. inhomogeneous system it does, and thus you
have to figure out what kind of meaning information you can derive from
that. typically, one would have to additional partitioning and also derive
expressions for looking at properties separately for the direction of
inhomogeneity and perpendicular to it.
this is all not new stuff. there were papers in the published literature on
issues like this when i started working on simulations of water at
interfaces way back in the mid 1990s. i suggest to dig into the literature
and *first* determine what is a meaningful property and interpretation
before even worrying about what a suitable command is and how to input it
to LAMMPS. just generating some data without knowing what it is going to
mean is a waste of time.
axel.