Dear LAMMPS users,
I’m doing a tensile test on a nanomembrane simulation box with ( P P S ) boundary condition. The thing is although I’m getting similar stress-strain curve for the different size of simulation box ( I’m replicating a unit cell in x and y ), for some size I’m observing floating atoms as I showed in the image attached. Could you please comment on how to solve it ? I’m asking because I have periodic in x and y and therefore I should observe same behaviour despite the number of duplication in those direction.
Thanks,
Ali
Does it happen after fracture?
Sanjib
It could be due to high strain rate. Moreover, you can’t expect that each model will behave exactly in the same manner.
Anyway, I think you don’t need to be worried about this unless you analysis progressive damage or post-fracture behavior.
Sanjib
Your structure is only 1D periodic - that is likely the problem lies.
Ray
Hi Ray,
Actually it is 2D periodic, i.e. ( p p s ).
Ali
Your structure is only 1D periodic - that is likely the problem lies.
No - by using “p p s” does not automatically give you 2D periodicity. Your structure is only 1D periodic.
Ray
And why is it surprising? In some cases, you have atoms that detach from the main sample. In other cases, not.
Is it dynamic simulation? If yes, try with different velocity seed to reproduce (or not) the result.
Is it quasi-statics? Try with a different minimizer.
I see. But, that seems strange to me as for some specific replication of unit cell in x and y I don’t observe floating atoms in z direction. Moreover, the stress-strain which is representative of the mechanical behavior is the same for both case.
Ali