Percolation of a rod-like particle in a static bed of spheres: trapping and passing

This study numerically demonstrates that the percolation of rod-like particles through a static bed of spheres is governed by a transition between trapping and passing regimes determined by rod length and pore geometry, where shorter rods move nearly twice as fast as longer ones due to reduced susceptibility to geometric trapping.

The Gist

Imagine a giant, invisible sieve made of thousands of large, smooth marbles packed tightly together. Now, imagine dropping a handful of different objects into this sieve: some are tiny marbles, and others are long, smooth sticks (like uncooked spaghetti or toothpicks).

This paper is a computer simulation that watches what happens when these "sticks" try to fall through the gaps between the big marbles under the pull of gravity. The researchers wanted to understand why some objects fall all the way through, while others get stuck.

Here is the story of what they found, broken down into simple concepts:

1. The Two Outcomes: The "Pass" and the "Trap"

When the sticks fall, they end up in one of two camps:

• The Passers: These sticks find a path, wiggle through the gaps, and fall all the way through the bed of marbles at a steady speed.
• The Trapped: These sticks fall for a while, but eventually, they get jammed. They stop moving and stay stuck inside the pile of marbles.

The paper discovered that whether a stick gets trapped or passes depends mostly on how long it is compared to the size of the gaps between the marbles.

2. The "Key in the Lock" Problem

Think of the gaps between the marbles as tiny, irregular doorways.

• Short sticks are like small keys. They can easily twist and turn to fit through almost any doorway. They fall fast because they don't get caught.
• Long sticks are like long, rigid pipes. To get through a doorway, a pipe has to be perfectly straight and aligned with the opening. If it hits the doorframe sideways, it gets stuck. Because the gaps in the pile are random and messy, long sticks frequently hit the "doorframe" at the wrong angle and get jammed.

3. The "Speed Limit" of Shape

The researchers found a surprising rule about speed: Shorter sticks fall almost twice as fast as longer sticks.

Why?

• Short sticks act almost like the big marbles themselves. They tumble easily and slide through the holes without much trouble.
• Long sticks have to do a lot of "dancing." As they fall, they have to constantly rotate to find a gap that fits their length. This constant twisting and turning slows them down. It's like trying to walk through a crowded room: a small child can dart through the crowd easily, but a tall person holding a long ladder has to stop, turn, and wait for a clear path, slowing their progress significantly.

4. The "Stuck" Moment

When a stick finally gets trapped, it doesn't just stop instantly like a car hitting a wall. It slows down over a very short distance (about half the width of one of the big marbles) before freezing.

The paper also looked at how they get stuck:

• Short sticks usually get caught standing up vertically, wedged between the sides of the marbles.
• Long sticks get stuck in all sorts of weird angles. They often get caught by touching three or four marbles at once, creating a complex "knot" that holds them in place.

5. The "Magic Number"

The researchers found a specific "tipping point." If a stick is longer than about half the width of the big marbles, it starts to have a high chance of getting trapped. If it's shorter than that, it almost always makes it through.

The Big Picture

The main takeaway is that shape matters just as much as size. In a world of round marbles, size is the only thing that determines if you fall through. But when you introduce long, thin shapes, the rules change. Being long makes you slower and much more likely to get stuck, not because you are heavy, but because you are hard to align with the messy, random holes in the pile.

This helps explain why, in nature or industry, long things (like fibers or grains) behave differently than round things (like sand or pills) when they are mixed together.

Technical Summary

Technical Summary: Percolation of a Rod-Like Particle in a Static Bed of Spheres

Problem Statement
Fine particle percolation through a static bed of larger particles is a fundamental mechanism in granular segregation, relevant to geophysical phenomena (e.g., landslides, avalanches) and industrial processes (e.g., pharmaceutical mixing, food processing). While the percolation of spherical particles has been extensively studied, real-world granular materials often contain non-spherical particles such as rods, fibers, or flakes. These anisotropic shapes introduce orientation-dependent constraints, interlocking, and distinct contact mechanics absent in spherical systems. This paper addresses the gap in understanding how rod-like particles percolate through a disordered static bed of larger spheres, specifically investigating the transition between continuous percolation and trapping, and how geometric anisotropy influences percolation velocity and dynamics.

Methodology
The study employs Discrete Element Method (DEM) simulations using the MercuryDPM code. The system consists of a static, disordered bed of 3,000 frictionless spheres (diameter $d_L$) with a packing fraction $\phi = 0.60$, prepared via isotropic compression and decompression to ensure a random, jammed state. Rod-like particles are modeled as chains of collinear, frictionless "glued" spheres (diameter $d_S$) with varying aspect ratios ($A = l/d_S$) and size ratios ($R = d_L/d_S$).

Key simulation parameters include:

• Rod Geometry: Rods are defined by the number of constituent spheres ($n$), aspect ratio $A$, and dimensionless length $L^* = l/d_L$. The study explores $1 \le A \le 10$ and $0.1 \le L^* \le 1$ across various $R$ values (including the critical geometric trapping ratio $R_t \approx 6.464$).
• Interaction Model: Normal forces follow a linear spring-dashpot model. Tangential forces are set to zero to isolate geometric effects. Rod-rod interactions are disabled; rods interact only with the static bed spheres.
• Protocol: 1,000 to 5,000 non-interacting rods are dropped into the bed under gravity. Trajectories are tracked to determine percolation depth, velocity, orientation, and contact statistics. The restitution coefficient is calculated internally based on the damping parameters, resulting in $e \le 0.6$ for sphere-sphere contacts.

Key Contributions and Results

1. Identification of Two Distinct Regimes:
   The study identifies a passing regime, where rods percolate continuously with a constant mean velocity, and a trapping regime, where rods stop after penetrating a limited distance. The transition between these regimes is governed by the rod length ($L^*$) and the size ratio ($R$).

   • A critical dimensionless length $L_c^* \approx 0.52$ (for $R=7$) marks the transition. Below this length, rods percolate indefinitely; above it, trapping probability increases with depth.
   • The characteristic penetration length $\lambda$ for trapped rods follows a power-law divergence as $L^*$ approaches $L_c^*$ from above: $\lambda/d_L \propto (L^* - L_c^*)^{-1}$.

2. Trapping Mechanisms and Geometry:

   • Contact Number: Trapped rods are immobilized with an average contact number $\langle Z \rangle = 5$. This is below the isostatic limit for spheres ($\langle Z \rangle = 6$) because, in the absence of friction, rods retain rotational freedom about their long axis. Trapping is driven by geometric entanglement rather than force balance alone.
   • Orientation: Short trapped rods ($L^* \approx 0.57$) are predominantly arrested in vertical orientations with contacts near the equator of the bed spheres. Longer trapped rods ($L^* = 1$) exhibit a broader distribution of orientations and contact locations (spanning from pole to equator).
   • Contact Location: For short rods, contacts are concentrated in the upper hemisphere ($20^\circ \le \theta \le 60^\circ$). As rod length increases, contacts spread toward the equator and lower hemisphere, reflecting increased geometric frustration.

3. Percolation Velocity Scaling:

   • The dimensionless percolation velocity $\tilde{v}_p = v_p / \sqrt{gd_L}$ decreases as the rod length ($L^*$) and aspect ratio ($A$) increase. Short rods percolate nearly twice as fast as long rods due to fewer geometric constraints.
   • The data for all rod geometries, including the single-sphere limit, collapses onto a single curve when scaled by $\sqrt{gd_L(R - R_p)}$, where $R_p \approx 4.5$ represents a characteristic pore diameter ratio. This scaling unifies the dynamics of spheres and rods.
   • Velocity is strongly orientation-dependent for longer rods; vertical alignment yields significantly higher velocities than horizontal alignment.

4. Kinetic Energy Distribution:
   Analysis of kinetic energy ratios ($K_{rot}^{rms}/K_{tra}^{rms}$) reveals that for short rods, rotational and translational energies are nearly equipartitioned. As rod length increases, rotational excitation is suppressed relative to translation due to the geometric constraints of the pore network restricting rotational freedom.

Significance
The paper demonstrates that particle shape anisotropy fundamentally alters granular percolation by introducing dynamical constraints and thresholds distinct from spherical systems. The findings reveal that elongation induces geometric frustration and enhances multi-point contacts, leading to reduced mobility and increased susceptibility to trapping. Specifically, the study establishes that:

• Rods do not behave as spheres; their ability to pass through pores is strictly coupled to orientation.
• A critical geometric threshold exists for trapping that depends on both rod length and the bed's pore structure.
• The percolation velocity of rods can be predicted using a scaling law derived from spherical percolation models, provided geometric constraints are accounted for.

These results provide a quantitative framework for predicting the transport and segregation of non-spherical particles in granular mixtures, with implications for understanding natural systems containing fibers or debris and optimizing industrial processes involving pills, grains, or fibrous additives. The work lays the foundation for extending granular segregation models to a broader range of anisotropic particle shapes.