I want to terminate some superficial atoms of a Ge slab with O, to do so I need to know the connections of this superficial atoms to assign the right position to the absorbed O, distance and orientation so to reproduce absorption positions.
well, you can identify those surface atoms from within LAMMPS using
compute coord/atom
i wouldn't know how you would access any bond information to construct
geometries in LAMMPS, so i am still not clear about how you could even
use topology information. but once you have identified atoms on the
surface, you can give them a different type, create a group and also
create bonds for only those atoms.
that said, i would not want to do all of that from inside LAMMPS, as
this is tedious and LAMMPS was not designed to be a molecular editor.
i would use some scriptable molecular visualization or editor
software, or already build the entire geometry from an external
program/script.
axel.
I am doing that externally in fact, using matlab scripts. I just wanted to figure out if there is any easy way to obtain the bonds topology with a manybody potential.
I am doing that externally in fact, using matlab scripts. I just wanted to figure out if there is any easy way to obtain the bonds topology with a manybody potential.
well, there is the create_bonds command and fix bond/create, which you
can use to create a bond topology based on certain heuristics and
rules. of course, you cannot do any meaningful computations after
that, but you can generate some kind of topology and write it out as a
data file, and then post process that kind of data.
i usually do such operations and scripting in VMD, which has tools
specific for molecules (but doesn't handle periodic systems very
well).
axel.