TL;DR
This paper presents a microscopic theory explaining why granular materials can exhibit "velocity-weakening" friction, a phenomenon responsible for instabilities like earthquakes and landslides. They demonstrate that this behavior, characterized by a non-monotonic friction law, emerges from endogenous mechanical noise within the material, even when individual particle contacts are perfectly stable.
What Problem Does This Paper Address?
Granular materials, like sand or rock fragments, are ubiquitous in nature and industry. A key characteristic is their macroscopic friction, which governs their flow behavior. While some materials exhibit "velocity-strengthening" friction (friction increases with flow rate, leading to stable flow), others show "velocity-weakening" friction (friction decreases with flow rate). This velocity-weakening is problematic because it can lead to sudden, catastrophic events such as stick-slip motion, landslides, and even earthquakes.
The Rice-Ruina rate-and-state model can describe these behaviors, but a fundamental microscopic understanding of why velocity-weakening occurs, especially in simple granular systems, has been lacking. Specifically, it was unclear how such instabilities could arise if the individual contacts between particles were themselves stable and didn't exhibit velocity-weakening. This paper aims to bridge this gap by developing a microscopic theory for velocity-weakening and the associated hysteresis (where the friction path differs depending on whether the flow rate is increasing or decreasing) in granular materials.
Methods
The authors employ a combination of theoretical modeling and numerical simulations using the Discrete Element Method (DEM).
- Theoretical Framework: They propose that "endogenous mechanical noise" – vibrations generated by particle collisions – plays a crucial role. This noise can "lubricate" the contacts, making them more prone to sliding. They develop a theory that links this mechanical noise to the fraction of sliding contacts and ultimately to the macroscopic friction law. A key aspect is identifying "soft spots" or "excitable regions" in the material, which they associate with rolling contacts on the verge of sliding.
- Numerical Simulations (DEM): They simulate 2D granular materials composed of frictional disks. Standard DEM techniques are used, including Coulomb friction at contacts (with identical static and dynamic coefficients), linear elastic-dashpot interactions between particles, and shear applied by moving top and bottom walls. They systematically vary parameters like particle deformability (represented by $\Delta$) and system size ($N$) to test their theoretical predictions.
Datasets
Not discussed in the paper. The authors generate their own numerical data through simulations.
Key Insights and Ideas
- Emergent Velocity-Weakening: Velocity-weakening friction is not a property of individual contacts but an emergent collective phenomenon arising from the interplay of many particles and their internal noise.
- The Role of Mechanical Noise: Endogenous acoustic noise, generated by particle collisions, is the key driver. This noise causes contacts that are close to sliding to do so intermittently.
- "Soft Spots" as Pre-Sliding Contacts: The theory identifies "soft spots" as rolling contacts that are just about to slide. The density of these spots follows a specific power-law distribution, $\propto df^{-\phi_s}$, with a non-trivial exponent $\phi_s \approx -1/3$.
- Positive Feedback Loop: Mechanical noise leads to more sliding contacts, which in turn leads to faster flow, more collisions, and thus more mechanical noise – a self-amplifying cycle.
- Hysteresis Explained: This positive feedback loop naturally explains the observed hysteresis in the friction law, where the friction is higher when increasing flow rate than when decreasing it.
- Deformability is Key: Increasing particle deformability (larger $\Delta$) reduces hysteresis and velocity-weakening. This is because softer particles have a larger potential energy per contact, making them less susceptible to lubrication by mechanical noise.
- Non-trivial Scaling: Several microscopic quantities (like the fraction of sliding contacts and sliding velocity) exhibit scaling behavior that depends on the dimensionless inertial number ($I$) and particle deformability ($\Delta$) in a non-trivial way, specifically as $I/\Delta^{1/4}$.
- Hard Particle Limit is Singular: Surprisingly, these quantities do not have a well-defined limit as particles become infinitely hard ($\Delta \to 0$). This suggests a singularity at the hard-particle limit for these properties.
Key Results
- Numerical simulations confirm the existence of a non-monotonic friction law ($\mu(I)$) for frictional particles, leading to hysteresis, especially for smaller systems and less deformable particles.
- Frictionless particles, in contrast, show negligible velocity-weakening and hysteresis, supporting the idea that friction is essential for this phenomenon.
- The theory predicts specific scaling laws for the fraction of sliding contacts ($x$), mechanical noise ($R$), and sliding velocity ($L_T$), which are found to be in good agreement with numerical results. For example, $x \propto (I/\Delta^{1/4})^{-\phi_s/(1+\phi_s)}$ and $R \propto (I/\Delta^{1/4})^2$.
- The theory predicts that external vibrations can eliminate hysteresis if they are strong enough, a finding consistent with experimental observations.
Limitations and Future Work
- Mean-Field Approximation: The theory relies on a mean-field approximation for the characteristic velocity of particle collisions, which might not be exact for frictional particles.
- Contact Mechanics Simplifications: The model uses harmonic contact descriptions, while real granular materials can have more complex Hertzian contact mechanics. The authors note that their framework can be extended to account for this.
- Frictionless Particle Hysteresis: While the paper focuses on frictional particles, they briefly mention that very soft and elastic frictionless particles might exhibit some hysteresis through other mechanisms.
- Future Work:
- Investigating the exponent $\phi_s$ for excitations in frictional packings.
- Extending the theory to over-damped suspensions.
- Exploring the link between anomalous compaction and mechanical noise.
- Experimental verification of the predicted scaling laws and the vanishing of velocity-weakening for highly deformable particles.
Summary updated