Zero-temperature Avalanche Criticality Governing Dynamical Heterogeneity in Supercooled Liquids

Using molecular simulations, this paper demonstrates that the temperature and system-size dependence of dynamical heterogeneity in supercooled liquids can be explained by a zero-temperature avalanche criticality framework.

The Gist

Imagine a crowded dance floor where everyone is trying to move, but the music is slowing down. At first, people can dance freely. But as the music gets slower (cooling down), the crowd starts to freeze up. Some people are still wiggling and dancing (mobile), while others are completely stuck in place (immobile). This mix of "moving" and "frozen" groups is called Dynamical Heterogeneity.

For years, scientists have been arguing about why these frozen and moving groups get bigger as the temperature drops. Is it because the whole floor is getting tighter? Or is there a specific rule governing how the "frozen" zones spread?

This paper proposes a new, exciting answer: It's like a snowball effect, or an avalanche.

Here is the breakdown of the research using simple analogies:

1. The Setup: The Supercooled Liquid

Think of a supercooled liquid (like honey that has been cooled but hasn't turned into a solid yet) as a giant, chaotic game of "Musical Chairs" played on a very crowded floor.

• The Players: The molecules (particles).
• The Goal: To move around.
• The Problem: As the room gets colder, it gets harder to move. Some players get stuck in a tight spot.

2. The Old Theory vs. The New Idea

Scientists have been trying to explain why the "stuck" groups get larger as it gets colder.

• The Old Idea: They thought it was just a gradual slowing down, like traffic getting worse and worse.
• The New Idea (This Paper): The authors suggest it works like dominoes or a landslide.
  • Imagine one person in the crowd finally manages to wiggle free.
  • Because they moved, they accidentally push their neighbor, who then pushes the next person.
  • This creates a chain reaction—a small "avalanche" of movement.
  • The paper argues that as the temperature drops, these avalanches don't just happen randomly; they follow a strict mathematical rule called Criticality.

3. The "Zero-Temperature" Avalanche

The term "Zero-Temperature" sounds scary, but think of it as "The Ultimate Tipping Point."

• Imagine a pile of sand. If you add one grain, nothing happens. Add another, nothing happens. But eventually, you reach a point where one single grain causes the whole pile to slide.
• The researchers found that in these liquids, the "avalanches" of movement behave exactly like that pile of sand right before it collapses.
• They discovered a specific "threshold temperature" (let's call it the Tipping Point). Above this temperature, the crowd moves somewhat randomly. Below this temperature, the movement becomes a structured, predictable avalanche.

4. How They Proved It (The Detective Work)

The scientists didn't just guess; they ran massive computer simulations (like a super-advanced video game) of millions of particles.

• The Measurement: They measured how "jumpy" the crowd was. They found that as the temperature dropped below the Tipping Point, the size of the "jumpy" groups grew in a very specific, predictable way.
• The Scaling: They tested different crowd sizes (small rooms vs. huge stadiums). They found that the rules of the avalanche were the same regardless of the room size. This is the "smoking gun" that proves it's a fundamental law of nature, not just a fluke.

5. The "Stokes-Einstein" Puzzle

There is a famous rule in physics (the Stokes-Einstein relation) that says: If something moves slowly, it should diffuse (spread out) slowly in a predictable way.

• The Problem: In supercooled liquids, this rule breaks. The molecules move slowly, but they spread out much faster than the rule predicts.
• The Solution: The avalanche theory explains this! Because the movement happens in big, sudden "bursts" (avalanches) rather than a slow, steady crawl, the particles get a "free ride" on these bursts. It's like a commuter stuck in traffic who suddenly gets a lift from a friend in a sports car—they move much faster than the traffic flow suggests.

The Big Takeaway

This paper changes how we view the "freezing" process. It's not just a slow, gradual stiffening. Instead, it's a critical transition where the system becomes hyper-sensitive. A tiny nudge can trigger a massive chain reaction of movement.

In a nutshell:
Cooling down a liquid is like building a snowdrift. As it gets colder, the snow becomes unstable. Eventually, you reach a point where the snow doesn't just sit there; it's ready to avalanche. The "moving" and "frozen" patches we see are just the visible signs of these microscopic avalanches happening over and over again. The authors have finally found the mathematical rulebook that governs these avalanches.

Technical Summary

1. Problem Statement

Supercooled liquids exhibit Dynamical Heterogeneity (DH), a phenomenon where mobile and immobile domains coexist on a mesoscopic scale. As the temperature ($T$) decreases, the characteristic size of these domains grows, and the dynamical susceptibility ($\chi_4$) exhibits significant system-size dependence.

• The Core Debate: While the existence of DH is well-established, its physical origin remains controversial. Existing theories include Mode-Coupling Theory (MCT), Random First-Order Transition Theory (RFOT), and structural order parameters.
• The Gap: Recent work on lattice models (thermal elastoplastic models) suggested that DH could be explained by zero-temperature avalanche criticality, analogous to plastic deformation in athermal glasses. However, it was unclear if this mechanism applies to particle-based systems (like molecular liquids) and how it relates to the system-size dependence of $\chi_4$.

2. Methodology

The authors employed Molecular Dynamics (MD) simulations to investigate the Kob–Andersen Model (KAM), a canonical binary Lennard-Jones mixture representing supercooled liquids.

• Simulation Setup:
  • System: 3D KAM with $N$ particles ($200 \le N \le 1500$) in the NVT ensemble.
  • Temperature Range: $0.44 \le T \le 1.0$ (above the MCT crossover temperature $T_{MCT} \approx 0.435$).
  • Potential: Smoothed Lennard-Jones potential to allow for saddle-point analysis.
• Key Observables:
  1. Dynamical Susceptibility ($\chi_4$): Defined via the overlap function $q(t)$ of A-type particles. The peak value $\chi^*_4$ quantifies the magnitude of DH.
  2. Correlation Length ($\xi$): Adopted from previous finite-size scaling analyses of the Binder parameter.
  3. Saddle-Point Analysis: Instead of analyzing instantaneous normal modes (which overestimate instability at finite $T$), the authors analyzed saddle-point configurations of the potential energy landscape. They calculated the fraction of unstable modes ($f^\dagger_{saddle}$) at these points.
  4. Displacement-Based Susceptibility ($\tilde{\chi}_4$): A modified susceptibility based on Mean Squared Displacement (MSD) to distinguish between lattice-based and event-based avalanche definitions.
• Scaling Analysis: The authors tested the finite-size scaling (FSS) hypothesis assuming a zero-temperature criticality picture:
  • $\xi \sim T^{-\nu}$
  • $\chi^*_4 \sim T^{-\gamma}$
  • $\chi^*_4 \sim \xi^{d_f}$ (where $d_f$ is the fractal dimension).

3. Key Contributions and Results

A. Validation of Zero-Temperature Avalanche Criticality

The study demonstrates that the temperature and system-size dependence of $\chi^*_4$ in the KAM can be fully explained by a zero-temperature avalanche criticality picture.

• Scaling Collapse: Using critical exponents determined independently (see below), the data for $\chi^*_4$ across different system sizes collapses onto a single master curve when plotted as $L^{-\gamma/\nu}\chi^*_4$ vs. $TL^{1/\nu}$.
• Critical Regime: This critical behavior is observed only below a threshold temperature $T_{ava} \approx 0.6$. Above this temperature, the scaling fails.

B. Independent Determination of Critical Exponents

The authors determined the critical exponents without relying on the susceptibility data itself, ensuring a robust test of the theory:

1. Correlation Length ($\nu$): By fitting the known correlation length $\xi$ (from Binder parameter analysis) to $T^{-\nu}$ in the regime $T \le T_{ava}$, they found $\nu \approx 3.2$.
2. Fractal Dimension ($d_f$) and $\gamma$:
   • They analyzed the fraction of unstable modes at saddle points, $f^\dagger_{saddle}$.
   • Theoretical scaling predicts $f^\dagger_{saddle} \sim T^{\nu(d - d_f)}$.
   • Simulation results showed $f^\dagger_{saddle} \sim T^{3.6}$.
   • Solving for $d_f$: $3.6 = \nu(d - d_f) \implies d_f \approx 1.9$.
   • Consequently, $\gamma = \nu d_f \approx 3.2 \times 1.9 \approx 6.0$.

• Consistency Check: Using $\nu \approx 3.2$ and $\gamma \approx 6.0$, the FSS collapse of $\chi^*_4$ was successful, confirming the avalanche criticality picture.

C. Identification of the Scaling Regime ($T_{ava}$)

The threshold $T_{ava} \approx 0.6$ was identified as the onset of stability enhancement:

• The authors analyzed the vibrational density of states (VDOS) of inherent structures.
• The prefactor $A_S$ of the low-frequency power law ($D_0(\omega) \sim A_S \omega^{6.5}$) decreases as $T$ drops below $0.6$.
• This decrease indicates an increase in mechanical stability, marking the transition where thermal avalanches governed by zero-temperature criticality become the dominant mechanism.

D. Breakdown of the Stokes-Einstein (SE) Relation

The paper explains the breakdown of the SE relation ($D\tau \sim \text{const}$) using the distinction between two avalanche size definitions:

1. Lattice-based ($S$): Corresponds to $\chi^*_4$ (spatial extent).
2. Event-based ($\tilde{S}$): Corresponds to $\tilde{\chi}_4$ (total number of relaxation events).

• The authors found distinct fractal dimensions: $d_f \approx 1.9$ (for $S$) and $\tilde{d}_f \approx 2.3$ (for $\tilde{S}$).
• The SE breakdown follows the scaling $D\tau_x \sim T^{\nu(d_f - \tilde{d}_f)}$.
• Crucial Finding: The SE relation holds only when the characteristic time scale $\tau_x$ is defined as the peak time of the event-based susceptibility ($\tilde{\tau}_4$), not the standard $\alpha$-relaxation time ($\tau_\alpha$). This resolves a previous inconsistency in the literature.

E. Rejection of MCT Criticality

The authors explicitly tested the Mode-Coupling Theory (MCT) hypothesis, assuming $T_{MCT} \approx 0.435$ as the critical point.

• Result: Finite-size scaling using MCT exponents failed to collapse the data.
• Conclusion: The observed DH growth is not driven by the MCT crossover but by the zero-temperature avalanche criticality starting at $T_{ava} \approx 0.6$.

4. Significance

• Unified Framework: The paper provides a consistent physical picture linking dynamical heterogeneity, system-size dependence, and the breakdown of the Stokes-Einstein relation in supercooled liquids to zero-temperature avalanche criticality.
• Mechanism Clarification: It identifies a specific temperature regime ($T \le T_{ava}$) where stability enhancement triggers this critical behavior, distinguishing it from the MCT regime.
• Methodological Advance: By utilizing saddle-point analysis and independent determination of exponents, the study offers a rigorous validation of the avalanche picture in particle-based systems, moving beyond lattice models.
• Future Outlook: The authors note that below $T_{MCT}$, the susceptibility saturates, suggesting a different mechanism governs deeply supercooled liquids, but the avalanche picture successfully describes the dynamics in the supercooled regime above $T_{MCT}$.

In summary, this work establishes that the complex dynamics of supercooled liquids near the glass transition are governed by critical phenomena analogous to zero-temperature plastic avalanches, providing a robust theoretical foundation for understanding dynamical heterogeneity.