Spatiotemporal Hierarchy of Slow Avalanches During Creep

1. Problem Statement

Amorphous solids (e.g., glasses, granular materials, crumpled sheets) far from equilibrium exhibit structural relaxations characterized by intermittent bursts of activity, known as avalanches. While traditional models describe these as scale-free, system-spanning events driven by mechanical instability, recent studies suggest that at finite temperatures, a new mechanism emerges: thermal avalanches.

In these systems, thermal noise facilitates the activation of local rearrangements. However, a complete description of the spatio-temporal structure of these thermal avalanches has been missing. Specifically, it is unclear how mechanical triggering (immediate, deterministic) and thermal activation (delayed, stochastic) interact to produce the observed slow, logarithmic creep and heavy-tailed temporal correlations (reminiscent of seismic aftershocks). The central problem is to disentangle these mechanisms and elucidate the hierarchical organization of relaxation events in disordered matter.

2. Methodology

The authors employed a multi-faceted approach combining advanced molecular dynamics simulations and experimental validation:

• Simulation Model:

  • They utilized a minimal structural model of an amorphous solid: a 2D disordered network of $\approx 10^5$ bi-stable elastic bonds.
  • Each bond possesses a double-well potential, representing two stable states separated by an energy barrier.
  • The network is over-constrained ($\Delta Z \approx 1.5$) and subjected to a constant external load ($F$) at a very low temperature ($T=0.001$).
  • Langevin Dynamics: Simulations were performed using LAMMPS with a Langevin thermostat to capture thermal fluctuations.
  • High-Resolution Tracking: A custom modification to LAMMPS allowed for the tracking of individual bond snap events with single-time-step accuracy over extremely long simulation times ($6 \times 10^6$ time units), capturing the full network configuration only upon instability.

• Data Analysis Strategy:

  • Event Classification: The authors analyzed the joint probability distribution of waiting times ($\Delta t$) and spatial distances ($\Delta r$) between consecutive snap events.
  • Hierarchical Grouping:
    1. Cascades: Events with short waiting times ($\Delta t < \Delta t^*$) and short distances ($\Delta r < R_c$) were grouped into "mechanical cascades."
    2. Avalanches: These cascades were then treated as point events. Cascades separated by longer, age-dependent waiting times but within a spatial correlation length were grouped into "thermal avalanches."
  • Temporal Scaling: Analysis was conducted in log-time ($\log t$) to account for the logarithmic aging of the system and to reveal power-law correlations.

• Experimental Validation:

  • The findings were validated against experimental data from the slow relaxation (creep) of crumpled Mylar sheets under constant load, where acoustic emission measurements identified discrete snap-trough instabilities.

3. Key Contributions

The paper makes several fundamental contributions to the understanding of non-equilibrium dynamics in disordered systems:

1. Disentanglement of Mechanisms: The authors systematically separated mechanical triggering (fast, age-independent, local) from thermal activation (slow, age-dependent, facilitated by stress redistribution).
2. Discovery of Spatio-Temporal Hierarchy: They revealed a two-level hierarchical structure:
   • Level 1 (Cascades): Fast, compact clusters of mechanically triggered snaps.
   • Level 2 (Avalanches): Slow, system-spanning sequences where cascades facilitate the delayed thermal activation of subsequent cascades via long-range elastic interactions.
3. Identification of Dual Power-Law Regimes: The study identified two distinct power-law regimes in the waiting time distributions, corresponding to intra-avalanche (within a thermal avalanche) and inter-avalanche (between avalanches) dynamics.
4. Robust Analysis Framework: They developed a "bottom-up" recipe for identifying avalanches in noisy, coarse-grained, or imprecise data, which is applicable to systems ranging from structural glasses to seismic faults.

4. Key Results

• Mechanical Cascades:

  • Dynamics: Occur at short timescales ($\Delta t < 10$) and are age-independent.
  • Structure: They are spatially compact (radius of gyration $R_g \sim S_c^{3/5}$, where $S_c$ is magnitude) and exhibit a power-law magnitude distribution ($P(S_c) \sim S_c^{-2}$).
  • Nature: Driven by immediate mechanical destabilization of neighboring bonds.

• Thermal Avalanches:

  • Dynamics: Occur at long timescales ($\Delta t > 10$) and are age-dependent (dynamics slow down as the system ages).
  • Structure: They are non-compact and system-spanning. The radius of gyration scales as $R_g \sim S_a^{1/2}$ (fractal dimension $d_f \approx 2$), indicating they cover the entire network.
  • Magnitude: The total number of snaps in an avalanche follows a power law $P(S_a) \sim S_a^{-2}$.

• Temporal Hierarchy & Correlations:

  • Intra-Avalanche: The waiting times between cascades within a single thermal avalanche follow a power law $P(\Delta \log t) \sim (\Delta \log t)^{-1}$.
  • Inter-Avalanche: The waiting times between distinct thermal avalanches follow a steeper power law $P(\Delta \log t) \sim (\Delta \log t)^{-2}$.
  • Significance: This dual power-law structure is mathematically analogous to the temporal correlations observed between earthquakes (Omori law and inter-event time distributions), suggesting a universal mechanism for triggered activity in disordered systems.

• Experimental Confirmation:

  • Analysis of acoustic emission data from crumpled Mylar sheets reproduced the same two-regime power-law behavior, confirming that the simulated hierarchy exists in real physical systems.

5. Significance

This work provides a unified framework for understanding slow relaxation in amorphous solids. By establishing that thermal avalanches are composed of mechanical cascades facilitated by long-range noise-mediated interactions, the paper bridges the gap between microscopic plasticity and macroscopic creep.

• Theoretical Impact: It challenges the view of avalanches as purely mechanical or purely thermal events, showing instead a complex interplay where mechanical bursts "prime" the system for delayed thermal activation.
• Broad Applicability: The identified hierarchy and dual power-law scaling offer a potential explanation for intermittent dynamics in diverse systems, including:
  • Seismology: Earthquake aftershock triggering.
  • Materials Science: Creep in metals, granular materials, and porous media.
  • Neuroscience: Potentially relevant to neuronal avalanche dynamics.
• Methodological Utility: The proposed analysis technique is robust against noise and low temporal resolution, making it a powerful tool for extracting hidden correlations in experimental data where direct tracking of individual events is impossible.

In conclusion, the paper elucidates the "rich structural dynamics" underlying the slow relaxation of amorphous solids, demonstrating that what appears as a continuous, slow creep is actually a hierarchical cascade of fast mechanical events and slow thermal facilitation.