I have a box of water molecules, and intend on inserting a cube of solid iron at the center of the simulation box, either by deleting water molecules in that region or by moving the cube from outside the box to the center. In the first case I find molecular bonds being removed during delete_atoms and delete_bonds operations, does not reset the ids of the atoms, bonds and angles for water, which gives an error. In the second option, it seems that moving the cube to the center will cause overlapping of atoms, causing an error. Is there a simple technique to set up this geometry. There was a post from Axel a while ago about using atomselect command in VMD. Some details on that or any suggestions would be helpful.
I have a box of water molecules, and intend on inserting a cube of solid
iron at the center of the simulation box, either by deleting water molecules
in that region or by moving the cube from outside the box to the center. In
the first case I find molecular bonds being removed during delete_atoms and
delete_bonds operations, does not reset the ids of the atoms, bonds and
lammps doesn't have the internal means to do this.
what you could do, is to first use the soft pair potential in
combination with fix adapt for your iron. you place
the iron atoms where you want them to be and slowly
grow the soft potential, to push the water molecules away.
you'd have to do this "staggered in shells" so that no water
gets trapped and you'd probably need to run with an npt
integrator to have the simulation cell expand as needed.
angles for water, which gives an error. In the second option, it seems that
moving the cube to the center will cause overlapping of atoms, causing an
error. Is there a simple technique to set up this geometry. There was a post
from Axel a while ago about using atomselect command in VMD. Some details on
that or any suggestions would be helpful.
yes this could be done. the vmd selection language is
very powerful and can easily carve out a suitable void
removing only complete water molecules. it is not easy
to describe this in a couple of sentences though. you
need to first familiarize yourself sufficiently with some
advanced VMD scripting. the vmd topotools package
has the necessary means to stitch your final system
together and renumber the atom ids as needed. but
it also may need some little scripting tweaks here and there.