Dear LAMMPS users,
In finite element simulations, for axisymmetric systems, only a quarter section needs to be modeled to save computational time.
Is it possible to use the same idea in molecular dynamics? For example, to simulate a nanowire, we can create a quarter of the nanowire in a cubic box. Let’s say the two symmetric planes of the nanowire are attached to the left and the bottom boundaries of the box, respectively. To maintain the axisymmetry, these two boundaries are still periodic but in a different way. For instance, a particle, moving down to the bottom of the box, can exit the boundary and re-enter the left side of the box with a velocity towards the right.
It seems that we can save lots of computational time by using quarter model for an axisymmetric system. But is it possible to implement this scheme in LAMMPS or is there any intrinsic difficulty associated with this idea?
Thanks and best,