I want to simulate a cutting process that involves two groups of atoms: one is named INDENTER, and the other is SAMPLE.
I have two questions:
1. If I am using "fix indent cylinder" command to simulate the INDENTER, how do I define a finite length of the cylinder? By the manual, the cylinder extends infinitely along its axis. Is there a way to modify that?
2. If I am using fix move command to define a group of atoms as the INDENTER, how do I define the interaction between the INDENTER and the SAMPLE? Such as friction coefficient, or the contact stiffness.
I want to simulate a cutting process that involves two groups of atoms: one is named INDENTER, and the other is SAMPLE.
I have two questions:
1. If I am using "fix indent cylinder" command to simulate the INDENTER, how do I define a finite length of the cylinder? By the manual, the cylinder extends infinitely along its axis. Is there a way to modify that?
you would have to modify the source, but if you want that kind of
detail, you are probably better off using option 2 anyway.
2. If I am using fix move command to define a group of atoms as the INDENTER, how do I define the interaction between the INDENTER and the SAMPLE? Such as friction coefficient, or the contact stiffness.
those would be implicit due to the way how you make the atoms
interact. if you are actually modeling at the atomic scale, those
properties have to be part of your model, i.e. the choice of surface
structure of your indenter and the interaction potential. this gets a
bit trickier with surface stiffness, there you'd have to assemble your
indenter from multiple layers where the inner layer is treated as a
rigid object and thus can be time integrated with fix move, but the
rest would have to move freely and thus - again - its properties would
be defined through the choice of interaction model and its parameters.
Note that there is a catch to the atomistic modeling of your indenter. If your aim is to simulate a non-rigid indenter and you go for option 2 you have to be careful with how large the size of the indenter is.
Elastic deformations tend to travel inside materials distances as large as the longest scale roughness of the surfaces in contact. If you go with option 2 you’ll have to ensure that the elastic deformations on the indenter as a result of compressing it onto the substrate die before reaching the uppermost layer that is to be treated as a rigid entity. Otherwise you’ll get reflection at that boundary thus compromising your results. For indentation lengths not very small you’ll be forced to include a lot of atoms in your model. Just keep this in mind as well as the fact that perhaps you can use the ATC package to reduce the number of DOFs of your problem without compromising accuracy.
Carlos
NB: I have zero experience with the ATC package therefore I am not sure if such hybrid simulations are fully doable within Lammps