[lammps-users] About periodic boundary condition

I want to use a moving top layer under PBC not non-PBC.
If ,however, I move top layer by addforce command or displace_atoms command
whatever, under PBC, the top layer atoms re-entered at the opposite side of
boundary.
This is problem because top layer should move toward shear force direction
in order to achieve shear deformation ( not to re-entered the box at the
opposite side)
This happens because PBC in LAMMPS forces atoms to remain in its original
box shape unless we use fix deform and fix displace_atoms (but both of them
can not apply constant shear force to the specimen). Why does LAMMPS have
this restriction. Can PBC in LAMMPS be periodic and shrink-wrapped ? (MDCASK
allow to do this and this doesn't makes any trouble, as far as i know)

You seem to be saying you want PBC, but you don't want atoms to leave
one side of the box and re-enter the other. I don't know what that means.
Can you explain more clearly what is the boundary/box constraints that
are needed?

Note that when you use displace_box (or fix_deform) there is an option to
change the box shape, but not adjust the atom coords to the new box. I don't
know if this is related to what you mean.

Also, what does it mean to apply constant shear-force to the specimen? Is
that the same as an off-diagonal pressure component? Which can be done
with fix npt.

Steve