rolling friction and cohesion models under development for granular models?

Hi, All,

I was looking through the materials of the fifth LAMMPS Workshop and Symposium and came across this presentation by Dan S. Bolintineanu:

http://lammps.sandia.gov/workshops/Aug17/pdf/bolintineanu.pdf

here it is indicated that complex contact potentials for granular simulations, including rolling friction and cohesion models are being added to new versions of lammps. My question is if there is an under development version of lammps with preliminary versions of this new features for public access (e.g. on GitHub) or this is still work in progress for private/(public) use and is not/(yet) released for public?

Thank you and have a great day.

Best wishes,
Bruce

Hi, All,

I was looking through the materials of the fifth LAMMPS Workshop and
Symposium <http://lammps.sandia.gov/workshops/Aug17/workshop.html> and
came across this presentation by Dan S. Bolintineanu:

http://lammps.sandia.gov/workshops/Aug17/pdf/bolintineanu.pdf

here it is indicated that complex contact potentials for granular
simulations, including rolling friction and cohesion models are being added
to new versions of lammps. My question is if there is an under
development version of lammps with preliminary versions of this new
features for public access (e.g. on GitHub) or this is still work in
progress for private/(public) use and is not/(yet) released for public?

this question is best directed at the author of the presentation (cc'd).​

axel.

Bruce,

These features are currently still under development. I expect we’ll have a working version of a more general granular pair style that includes rolling friction and cohesion, which we can add to the general release in a few weeks, maybe a couple of months at the latest.

Best,

Dan

Hi, Dan,

Thank you for your reply and information! Great news! I am happy to help with testing and evaluations, if it helps/is needed.

Best wishes,
Bruce

Here’s a cohesive model that I used some time ago to simulate rough (non-compliant) micron sized particles. It was adapted from pair_colloid and admittedly is in need of some clean up. This would essentially be the Derjaguin, Mueller, Toporov (DMT) model seen often in literature. Its also used by others in the chemical engineering literature (see refs in paper below). The model does not include granular interactions but can be used in tandem with them via pair_style hybrid. The input is as follows (resembling pair_style colloid)

dmt Hamaker_constant interatomic_distance contact_radius1 contact_radius2 outer_cutoff

Note that the contact radius may be different from the diameter if you are looking to incorporate surface roughness of a particle.

There are some numerical details about proper time-steps to use for energy conservation in collisions in the paper when combining with Hertz and Hooke models in LAMMPS. Murphy, E. and Subramaniam, S. “Binary collision outcomes for inelastic soft-sphere models with cohesion.” Powder Tech. 305 (2017) 462-476. (http://www.sciencedirect.com/science/article/pii/S0032591016305769) Also some hints if you’d like to speed things up, depending on your flow set-up.

Thought this could be of use for folks to use in the short term, while waiting on the Sandia folks. I’m also open to folks cleaning it up and incorporating it into LAMMPS if that is desired - meant to do it a long time ago myself.

HTH.

pair_dmt.cpp (18.6 KB)

pair_dmt.h (2.12 KB)