Out-of-equilibrium percolation transitions at finite critical times after quenches across magnetic first-order transitions

This paper demonstrates that quenching two-dimensional Ising-like systems across their magnetic first-order transitions induces an out-of-equilibrium percolation transition at a finite critical time, where the resulting critical behavior exhibits random-percolation scaling for cluster geometry but features a field-dependent exponent and a spinodal-like exponential dependence on the magnetic field strength.

The Gist

Imagine a crowded dance floor where everyone is holding hands. Half the people are wearing Red shirts, and the other half are wearing Blue shirts.

In this paper, the scientists are studying what happens when you suddenly change the rules of the dance floor.

The Setup: A Frozen Standoff

Imagine the dance floor is in a state of "frozen" balance. Everyone is holding hands with people wearing the same color. You have huge, solid blocks of Red people and huge, solid blocks of Blue people. They are stuck in a standoff.

In physics terms, this is a First-Order Transition. It's like water that is super-cooled but hasn't turned to ice yet. It's stable, but it's waiting for a nudge to flip to the other side.

The "Quench": The DJ Changes the Music

Suddenly, the DJ (the scientists) changes the music. They introduce a new rule: "Everyone who is Blue must now try to become Red!"

This is called a quench. The system is no longer in equilibrium; it's in a panic. The Blue people want to stay Blue (because they are used to it), but the new rule is pushing them to turn Red.

The Drama: The Battle of the Clusters

At first, the Red people are just a few scattered individuals. The Blue people are still in massive, dominant groups.

But as time passes, something magical happens. The Red people start finding each other. They grab hands, merge into small groups, and then those small groups crash into each other to form giant Red super-clusters.

At a very specific moment in time (let's call it The Critical Time), a massive shift occurs.

• The giant Blue block suddenly shatters.
• The scattered Red groups fuse together into one giant, all-encompassing Red wave that covers the entire floor.

This moment is the Percolation Transition. In everyday language, it's the exact second when the "Red" team takes over the whole room, and the "Blue" team is reduced to tiny, isolated islands.

What the Scientists Discovered

The researchers (Andrea, Davide, and Ettore) watched this process on a computer simulation of a 2D grid (like a giant checkerboard). They found some fascinating things:

1. The Shape of the Takeover
When the Red team finally takes over, the shape of their giant cluster looks exactly like the shapes you see in standard, random games of chance (like pouring water on a sponge). The "fractal dimension" (a fancy way of describing how jagged and complex the edge of the cluster is) is the same as if the Red people had just appeared randomly.

• Analogy: Even though the Red people were fighting a battle, the shape of their victory looked like a random splash of paint.

2. The Speed of the Takeover
Here is where it gets weird. While the shape of the takeover looked random, the speed at which it happened was very different from a random game.

• In a random game, the transition happens at a predictable speed.
• In this "fighting" scenario, the speed depends entirely on how strong the DJ's new rule was (the magnetic field, $h$).
• If the rule is weak, the Red team takes a long time to organize. If the rule is strong, they take over quickly.
• Analogy: Imagine a rumor spreading. If the rumor is weak, it spreads slowly. If the rumor is explosive, it spreads instantly. The "explosiveness" here depends on the strength of the external push.

3. The "False Vacuum" Effect
The paper suggests that the Red people don't just grow one by one. Instead, they act like a swarm of bees. Small Red groups form, and then they merge with other small groups to become huge.

• Analogy: It's not like a single giant tree growing from a seed. It's like a forest fire where small patches of flame merge to create a massive wall of fire that consumes everything. This merging process is what allows the system to switch phases so quickly, faster than if they had to wait for a single giant "perfect" group to form.

Why Does This Matter?

Usually, scientists study "equilibrium" (things at rest) or "explosive" transitions that require weird, non-physical rules (like forcing people to only hold hands with specific neighbors).

This paper shows that real, physical systems (like magnets) naturally do this. If you push a magnet hard enough across a tipping point, it doesn't just slowly change; it undergoes a sudden, dramatic "percolation" event where the new state suddenly takes over the whole system.

The Takeaway

Think of it like a political election in a town that is perfectly split 50/50.

• The Old View: We thought the new party would slowly win people over one by one.
• The New View: The paper shows that once the new party gets a tiny bit of momentum, they don't just win voters; they win entire neighborhoods at once. Suddenly, the map flips from mostly blue to mostly red in a single, dramatic moment.

The scientists have mapped out exactly when that flip happens and how the shapes of the winning groups look, revealing that nature has a very specific, predictable way of flipping from one state to another when pushed out of balance.

Technical Summary

1. Problem Statement

The paper investigates the non-equilibrium dynamics of ferromagnetic systems driven across a magnetic first-order phase transition. Specifically, it addresses the question of whether a percolation transition occurs during the relaxational evolution of a system quenched from a metastable phase to a stable phase.

While standard percolation theory (Random Percolation, RP) is well-understood for static systems, and "explosive percolation" (EP) has been studied in network models with non-local, irreversible rules, this work explores a physically realizable scenario: a local, reversible, relaxational dynamics (heat-bath dynamics) in a standard statistical mechanical model (the 2D Ising model) undergoing a sudden quench of the magnetic field across a first-order transition line. The authors aim to characterize the geometric properties of the spin clusters during this transition and determine if they follow standard RP universality or exhibit new critical behaviors.

2. Methodology

The authors employ numerical simulations and finite-size scaling (FSS) analysis on the two-dimensional (2D) nearest-neighbor Ising model on a square lattice with periodic boundary conditions.

• Model Hamiltonian:
  $$H(\beta, h) = -\beta \sum_{\langle x,y \rangle} s_x s_y - h \sum_x s_x$$
  where $s_x = \pm 1$, $\beta > \beta_c$ (low temperature), and $h$ is the external magnetic field.
• Quenching Protocol:
  1. The system is initialized in equilibrium at $h_i < 0$ (negatively magnetized phase) with $\beta > \beta_c$.
  2. At $t=0$, the magnetic field is suddenly quenched to a positive value $h > 0$.
  3. The system evolves under single-spin heat-bath dynamics (a local relaxational dynamics) with the new field $h$.
• Observables:
  • Magnetization: $M(t) = \langle \sum s_x \rangle / V$.
  • Cluster Sizes: The sizes of the largest connected clusters of positive ($S_+$) and negative ($S_-$) spins are tracked using the Hoshen-Kopelman algorithm.
  • Rescaled Sizes: $\Sigma_{\pm}(t) = \langle S_{\pm} \rangle_t / V$.
  • Ratio: $R_s(t) = \langle S_- \rangle_t / \langle S_+ \rangle_t$.
• Analysis:
  • Simulations were performed for various lattice sizes ($L$ up to 4000) and field strengths ($h \in [0.045, 0.07]$).
  • Finite-size scaling (FSS) ansatzes were applied to determine critical times ($t_c$) and critical exponents.
  • The infinite-volume magnetization $M_\infty(t)$ was extracted by analyzing the convergence of data as $L \to \infty$.

3. Key Contributions and Results

A. Emergence of a Dynamic Percolation Transition

The study reveals that at a finite critical time $t_c(h)$, the system undergoes a sharp out-of-equilibrium percolation transition.

• Mechanism: The transition is marked by the simultaneous percolation of the largest positive-spin cluster and the inverse percolation (disappearance) of the largest negative-spin cluster.
• Visual Evidence: Snapshots show that around $t_c$, positive clusters rapidly merge to form a single spanning cluster, while the large negative domains fragment. This occurs without the independent growth of isolated droplets; rather, it is driven by the collective aggregation (merging) of droplets.

B. Finite-Size Scaling (FSS) Behavior

The authors establish that the transition exhibits standard FSS behavior characteristic of percolation, but with specific nuances:

• Scaling Ansatz: The rescaled cluster sizes follow $\Sigma_{\pm}(t, L) \approx L^{-\delta} F_{\pm}((t-t_c)L^w)$.
• Fractal Dimensions ($d_{\pm}$): The fractal dimensions of the critical clusters are found to be equal ($d_+ = d_-$) and consistent with the standard Random Percolation (RP) value:
  $$d_{RP} = \frac{91}{48} \approx 1.896 \implies \delta = d - d_{RP} \approx 0.104$$
  This suggests the geometric universality class of the critical clusters is identical to standard RP, regardless of the magnetic field $h$.

C. Non-Standard Critical Exponent ($w$)

While the geometric properties match RP, the dynamical approach to criticality differs significantly:

• The exponent $w$, which controls the scaling of the time window near $t_c$, depends on the magnetic field $h$.
• $w$ is strictly less than the RP value ($w_{RP} = 3/4$) and decreases as $h$ decreases.
• The data suggests $w \propto h^\theta$ with $\theta \approx 0.64$. As $h \to 0$, $w \to 0$, implying the time window for the transition broadens and the dynamics slow down.
• Interpretation: This indicates a line of fixed points in the renormalization group sense, where metric exponents (like $\delta$) are universal, but the scaling of the approach to criticality ($w$) varies with the control parameter $h$.

D. Spinodal-like Scaling and Critical Time

The critical time $t_c(h)$ exhibits a spinodal-like exponential dependence on the field $h$:
$$t_c(h) \sim \exp\left(\sqrt{\frac{\sigma^*}{h}}\right)$$

• The magnetization $M_\infty(t)$ scales with the variable $\sigma = h(\ln t)^2$.
• The transition occurs at a critical value $\sigma^* \approx 3.03$.
• This behavior implies that the transition is not driven by the nucleation of a single stable critical droplet (which would require time $\sim \exp(1/h)$), but by the aggregation of sub-critical droplets. The merging process is energetically favored and allows the system to switch phases much faster than isolated droplet nucleation would predict.

4. Significance and Implications

1. New Class of Non-RP Percolation: The paper identifies a new type of non-Random Percolation transition that arises naturally in physical systems governed by local, reversible dynamics. This contrasts with "explosive percolation" models, which typically rely on non-local or irreversible ad-hoc rules.
2. Universality Decoupling: It demonstrates a decoupling between geometric universality (fractal dimensions matching RP) and dynamical universality (exponent $w$ depending on $h$). This challenges the assumption that the entire critical behavior must be identical to the static RP model.
3. Mechanism of First-Order Transitions: The results provide deep insight into the mechanism of phase transitions in 2D Ising systems. They confirm that the transition from metastable to stable phases is driven by collective aggregation rather than independent droplet nucleation, explaining the specific time scales observed.
4. Generality: The authors suggest this behavior is generic for systems with discrete symmetries undergoing first-order transitions, though the specific pattern (simultaneous vs. separate percolation) may differ in 3D systems (where a coexistence phase is observed).

Conclusion

The paper establishes that quenching a 2D Ising system across a first-order transition induces a dynamic percolation transition at a finite critical time. While the critical clusters share the fractal geometry of standard random percolation, the dynamical scaling exponents and the critical time itself exhibit a unique dependence on the driving field, governed by spinodal-like scaling laws and collective aggregation dynamics. This work bridges the gap between equilibrium percolation theory and non-equilibrium phase ordering kinetics.