[lammps-users] Harmonic bonds with random bond strength.

Dear all,

I would like to simulate a fracture process between two plates that are "glued" together. It is modeled this way: I have a system of atoms that interact with harmonic bond potentials, using the "bond_style harmonic" command. However, for some atoms (along the plane where the "glue" is), I want these harmonic bonds to break at a critical separation R_c. Is there already a potential that does this, or must I make my own version of an harmonic potential to do this?

Also, what I further would like is to have a _random_ distribution of critical separation values, so that the bonds along the "glued" plane will break at different loads/separations. Is there a way to do this with an harmonic bond type? Do anyone have suggestions to better ways of doing this?

Christer H. Ersland.

Look at fix bond/break. Also bond_style quartic. Both allow bonds to
break in different ways. If you want random Rc lengths, then you would
have to use one of these as a starting point, and do something more
general. I.e. the bond itself would need to store an input param, like
the randomized Rc, that it would use to determine breakage.