I am not sure this is the right place to ask this question but still here it is.
I run a simple LJ bead-spring simulation of linear polymer chains with breakable bonds and additional breakable crosslinks under p p p boundary conditions with about 20k beads. I apply constant force at z direction and zero force at the other directions. The sample undergoes uniaxial deformation. It all seems fine on appearance, but…
I cant get brittle fraction no matter how low the temperature is! This would look like the sample separates into two parts without a long neck. However, what I see is that there is always a neck and the picture doesnt change qualitatively regardless the temperature being far above or far below Tg=0.6epsilon(in this case).
Is there a way to augment the model to get that brittle fracture thing?
Its not only the temperature that has an effect on the fracture behavior. Strain Rate is also an equally important one. Trying to capture this macroscopic response( brittle fracture, ductile tearing) in MD requires the strain rates also to be in the limits. This might be one of the reason of your problem.
Another obvious reason may be the size effect. Of course with PBC, you can simulate as large a system you want but that doesnot mean you dont have to check for the artifacts introduced with such type of boundary condition. Check this part.
One of your issues is the properties of ‘Lennard-Jonesium’. This material best simulates ideal gas, and does not simulate metals at all well.
There are some fcc metals (Al, Ag) which (when very pure) will not undergo brittle cleavage fracture, even under very cold conditions or at very high strain rates.
It may be that Lenndard-Jonesium does not display brittle fracture; a quick scan of some ‘google’ results show that you have to play with the model to get cracking behaviour. So the fundamental answer to your last question is ‘yes’, but you will have to go look in the literature.