How do force chains in flowing sand become rigid?
In slowly sheared granular material, the network of strong contact forces becomes connected in exactly the way predicted by rigidity percolation theory, but faster flow changes the critical behaviour.
Source
Emergence of rigidity percolation in flowing granular systems
Study at a glance
- Design
- Computational / modelling — Molecular dynamics (LAMMPS) of two-dimensional frictionless bidisperse disks under simple shear, with percolation analysis of force networks and finite-size scaling across system sizes and shear rates.
- N
- No participant N; system sizes range from 2048 to 65,536 disks, each averaged over 10 independent simulations at each of several shear rates.
- Population
- Simulated dense (above jamming) packings of frictionless bidisperse disks in two dimensions
- Outcome
- Critical exponents (nu, beta, gamma, eta) and fractal dimension of the percolating interparticle-force network as a function of shear rate
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The authors simulated a dense two-dimensional packing of frictionless disks of two sizes being continuously sheared, using system sizes from 2048 to 65,536 particles and shear rates spanning three orders of magnitude. For each configuration they kept only contacts whose force exceeded a threshold and asked at which threshold a system-spanning cluster of force chains appears. Using finite-size scaling, they extracted the critical exponents of this percolation transition and compared them with known universality classes, and interpreted the shear-rate dependence with the extended Harris criterion for correlated disorder.
What they found
At the slowest shear rates the exponents (for example nu of about 1.21, beta of about 0.20 and gamma of about 2.15) matched standard rigidity percolation, which the authors describe as the first verification of this class in an off-lattice molecular dynamics simulation. Above a crossover shear rate the correlation-length exponent rose steadily, so flow rate acts as a relevant perturbation producing a continuous line of exponents, passing through random-percolation values at intermediate rates. The force-force correlation exponent and the fractal dimension of the spanning cluster, however, stayed essentially the same across all flow rates.
The limits
What it doesn't show
The system is idealised: two-dimensional, frictionless, athermal disks with simple linear contact forces, so the results may not carry over directly to frictional sand, three-dimensional flows or thermal glasses, which the authors flag as open questions. Percolation is defined by an arbitrary force threshold on contacts rather than by a direct rigidity (pebble-game) test of each cluster. The explanation of changing exponents via long-range flow-induced correlations is a theoretical argument fitted to the data, and the beta exponent had error bars too large for a conclusive test.
Key terms
- Percolation transition
- The point at which randomly added (or removed) links first create a cluster spanning the whole system.
- Rigidity percolation
- A stricter percolation in which the spanning cluster must be mechanically rigid, not merely connected.
- Critical exponent
- A number describing how a quantity such as correlation length diverges near a phase transition; shared exponents define a universality class.
- Finite-size scaling
- A method that uses how results change with system size to extrapolate critical properties to an infinitely large system.
- Force chains
- Filament-like paths of strongly loaded contacts that carry most of the stress in a granular packing.
- Harris criterion
- A rule that predicts whether added disorder (short- or long-range correlated) changes the critical exponents of a transition.
Flashcards
0 of 10 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
What kind of particles were simulated?
Common questions
Why vary the system size so much?
Exponents measured in a single finite system depend on its size; comparing many sizes lets researchers extrapolate to the infinite-size limit where the universality class is defined.
What does it mean that shear rate is a 'relevant perturbation'?
Changing it actually changes the critical exponents, rather than just shifting where the transition occurs, so faster flow puts the system in a different class.
Why might this explain disagreements in earlier studies?
If compression rate acts like shear rate, different preparation speeds in earlier jamming studies could have produced the range of conflicting exponents they reported.
More on Granular matter