regional pressure

Dear lammps users,

In my simulation I compress a nanopillar and the boundary atoms are fixed(fix 1 boundary setforce 0.0 0.0 0.0), so in order to compute the stress in the compressive direstion(say x), only mobile atoms(group mobile) need to be considered. I can use compute stress/atom command and then use compute reduce command(compute 1 mobile reduce sum c_stress[1]) to get the total x-diretion stress of the group, but how can I get the total volume of atoms in the mobile group? Since press equal stress divided by volume in lammps.

Thank you very much!

Michelle

Dear lammps users,

In my simulation I compress a nanopillar and the boundary atoms are fixed(fix
1 boundary setforce 0.0 0.0 0.0), so in order to compute the stress in
the compressive direstion(say x), only mobile atoms(group mobile) need to
be considered. I can use compute stress/atom command and then use compute
reduce command(compute 1 mobile reduce sum c_stress[1]) to get the total
x-diretion stress of the group, but how can I get the total volume of atoms
in the mobile group?

‚Äčthis is the million dollar question. what is the volume of an atom or a
group of atoms is not a well defined entity. if you search the mailing list
archives‚Äč, you'll find numerous discussions on the subject. for confined
atoms (i.e. atoms fully surrounded by other atoms) in a regular solid or
liquid, you can use the voronoi/atom compute as an estimate, but that will
fail badly for all atoms at a surface. another option might be to look at
the USER-ATC package, which has several tools to translate atomistic
properties to a finite element grid.

axel.