Emergent long-tail dynamics in driven magnets with dynamical frustration

arXiv:2511.09401 · cond-mat.stat-mech, cond-mat.str-el · Submitted 2025-11-12 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: We're looking at the paper: "Emergent long-tail dynamics in driven magnets with dynamical frustration".

Paper summary: Kai: So we're talking about this paper, "Emergent long-tail dynamics in driven magnets with dynamical frustration." The main thesis is that you can spontaneously get dynamical frustration to show up in systems that don't naturally have it when you drive them periodically >

Mira: It shows how a classical spin system can develop spin-ice physics under these conditions specifically when the drive-induced heating isn't too strong >

Kai: The paper focuses on magnetic monopole excitations, and they demonstrate that these aren't like their equilibrium versions; they follow a non-ergodic random walk process with waiting times that have long-tailed power laws >

Lev: Long-tailed means the waiting times don't decay exponentially; they have these power law distributions where the exponent itself can be tuned by the system’s effective temperature >

Mira: And they link this anomalous behavior directly to a drive-induced short-range attractive interaction between these monopoles that traps them upon collision >

Kai: So essentially, it's a drive-induced interaction that creates an energy barrier, causing them to stick together for longer than usual >

Lev: That trapping effect leads to the long-tail distribution of the duration of those plateaus, P(tau) equals tau minus eta where eta is greater than one >

Mira: And they found that this attraction is caused by higher-order terms in the drive that involve local multi-spin interactions resulting in these effective short-range interactions between the monopoles >

Kai: And they also detail how heating kicks in at intermediate driving frequencies, causing a plateau and then rapid heating to an infinite temperature state >

Lev: That initial plateau duration tau zero is interesting because it suggests a strong dependence on the initial conditions of the system when you're looking at this specific drive regime >

Mira: And they also show that this collision timescale for their two-dimensional random walk scales with L squared where L is the system size >

Kai: It really shows that these dynamics are tied to how quickly things are driven, and it’s not just a static property of the underlying lattice structure >

Conclusion: Kai: So looking at "Emergent long-tail dynamics in driven magnets with dynamical frustration," it seems like the big idea is that periodic driving can fundamentally change how a system behaves by making frustration appear where it wasn't supposed to be >

Mira: It’s about finding new physics in non-equilibrium systems, showing how simple driving can lead to complex, slow dynamics characterized by these long-tail waiting times for excitations >

Kai: The implication is that we need to look beyond equilibrium descriptions when studying driven magnets because the driving itself can induce these kinds of traps and slow relaxation processes >

Lev: For us in error correction research, this suggests that even if you suppress the heating, you still have to account for these kinds of anomalous trapping events during the evolution of your system >

Mira: And it opens up possibilities where collective behaviors could emerge between excitations that we don't see in simpler models because they are governed by these drive-induced attractions >

Kai: So ultimately, this paper suggests that the way we apply external driving to a magnet is just as important as the static magnetic interactions when you want to understand the resulting dynamics >

Chenyue Guo, *Hongzheng Zhao, *Zi Cai

State Key Laboratory of Artificial Microstructure and Mesoscopic Physics, Peking University · Wilczek Quantum Center and Key Laboratory of Artificial Structures and Quantum Control, Shanghai Jiao Tong University

cond-mat.stat-mech, cond-mat.str-el

Submitted: 2025-11-12

Updated: 2025-11-12

Journal ref: Proc. Natl. Acad. Sci. U.S.A. 123 (46) e2611202123, (2026)

DOI: 10.1073/pnas.2611202123

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 100/100

The gist: Emergent long-tail dynamics in driven magnets with dynamical frustration shows that dynamical frustration can spontaneously emerge in frustration-free magnetic systems under periodic driving…

Key concepts

Dynamical Frustration
Frustration, which usually means conflicting interactions in a system, can appear spontaneously in systems driven periodically. This happens because the driving process creates an effective time-averaged Hamiltonian that introduces frustration even when the instantaneous Hamiltonian is frustration-free. It's an emergent property of non-equilibrium driving.
Magnetic Monopole Excitations
These are specific excitations within a magnetic system that behave like particles. The paper tracks their movement by measuring spin flips and locating them in plaquettes. These excitations are key to understanding the anomalous dynamics, as their interactions lead to the observed long-tail waiting times.
Long-Tailed Waiting Times
The time between events for these magnetic monopoles does not follow a normal distribution; instead, it follows a power law. This means there is a high probability of very long waiting periods. This unusual behavior is due to an energy barrier created when two monopoles collide and stick together.
Drive-Induced Attraction
The periodic driving introduces higher-order terms in the Hamiltonian that create an effective short-range attractive force between magnetic monopoles. This attraction traps the monopoles after a collision, causing them to remain bound for extended periods before moving freely, which is responsible for the long-tailed waiting times.

Terminology

Summary

Emergent long-tail dynamics in driven magnets with dynamical frustration shows that dynamical frustration can spontaneously emerge in frustration-free magnetic systems under periodic driving Specifically, we consider a classical spin system and demonstrate the emergence of spin-ice physics when drive-induced heating is well suppressed In particular, we focus on the dynamics of magnetic monopole excitations, which exhibit a non-ergodic stochastic random-walk process with longtailed, power-law–distributed waiting times where the power-law exponent is tunable by the system’s effective temperature We show that a drive-induced short-range attractive interaction between monopoles is responsible for the long-tailed distributions observed in both monopole and heating dynamics. p. 0

Where the paper says it

Specifically, we consider a classical spin system and demonstrate the emergence of spin-ice physics when drive-induced heating is well suppressed. In particular, we focus on the dynamics of magnetic monopole excitations, which, in sharp contrast to their equilibrium counterparts, exhibit a non-ergodic stochastic random-walk process with longtailed, power-law–distributed waiting times, where the power-law exponent is tunable by the system’s effective temperature.

Emergence of Dynamical Frustration

Frustration itself could be an emergent property tied to the nonequilibrium feature of periodic driving By staggering competing interactions, we implement a timedependent Hamiltonian that is frustration free at any given time. When this periodic switching is too fast to be followed by the system, it will experience a time-averaged Hamiltonian where frustration dynamically emerges. p. 0

Where the paper says it

Therefore, frustration plays no role if the switching between the competing interactions is sufficiently slow, where the system keeps evolving under the frustration-free instantaneous Hamiltonian. In contrast, when this periodic switching is too fast to be followed by the system, it will experience a time-averaged Hamiltonian, where frustration dynamically emerges.

A generic Floquet many-body system generally heats up towards an infinite temperature state that is entirely featureless. p. 0
Where the paper says it

To address this question, for concreteness, we consider a Floquet protocol as shown in Fig. A generic Floquet many-body system generally heats up towards an infinite temperature state that is entirely featureless.

Monopole Dynamics and Long-Tail Distributions

The monopole excitations exhibit a non-ergodic stochastic random walk characterized by a long-tailed, power-lawdistributed waiting time with a divergent average where the power-law exponent is tunable by the background effective temperature Teff. This anomalous behavior is attributed to the drive-induced short-range interaction which introduces an energy barrier that traps two monopoles upon collision, thereby leading to an extended waiting time before their free motion resumes. p. 0

Where the paper says it

This is sharply distinct from the monopole dynamics in the undriven SI, where a standard random walk is generally expected [35]. We attribute this anomalous behavior to the drive-induced short-range interaction, which introduces an energy barrier that traps two monopoles upon collision, thereby leading to an extended waiting time before their free motion resumes.

The probability density of the duration of these plateaus is observed to follow a power-law decay P(τ) ∼ τ−η with a non-universal power η > 1. p. 2
Where the paper says it

As shown in Fig. 2 (c), for those long-lived plateaus, P(τ ) ∼ τ −η+1 is observed suggesting a long-tail power-law distribution p(τ ) ∼ τ −η with a non-universal power η > 1.

This long-tail distribution arises due to the energy barrier V induced by their attractive force, that the two monopoles must overcome before moving freely. p. 3
Where the paper says it

1J for (a)-(c). This long-tail distribution arises due to the energy barrier V induced by their attractive force, that the two monopoles must overcome before moving freely.

Heating Dynamics Under Different Driving Regimes

Heating becomes noticeable at intermediate driving frequencies. In the intermediate frequency regime (T∼J−1) where heating is accelerated, the stroboscopic energy εn experiences a plateau with duration τ0, after which it quickly increases to its infinite temperature value. p. 3

Where the paper says it

As shown in Fig. 3 (a), for different trajectories starting from initial states with various noise realizations, the stroboscopic energy εn experiences a plateau with duration τ0, after which it quickly increases to its infinite temperature value.

This suggests a strong initial-state dependence of the heating time in contrast to generic Floquet many-body systems. p. 3
Where the paper says it

3 (b), we plot P(τ0)= R ∞ τ0 p(τ ′ 0 )dτ ′ 0 , where p(τ0) is the probability density of τ0, which also decays algebraically with τ0 for large τ0, indicating again, a long-tail distribution. This thus suggests a strong initial-state dependence of the heating time, in contrast to generic Floquet many-body systems, where heating occurs on a typical time scale that depends only on the driving protocol and initial-state temperature [34].

The characteristic time scale for their collision during this 2D random walk scales as L2. p. 4
Where the paper says it

Before heating begins, the two monopoles wander through the lattice, and the typical timescale for their collision during this 2D random walk scales as L 2 , which explains the finite-size dependence of P(τ0). Conclusion and outlook – In summary, we propose a dynamical frustration mechanism in periodically driven systems, which exhibit a long-tailed, power-law distribution of the waiting time for monopole random walks.

Physical Interpretation and Future Directions

The emergent short-range attractive interaction can be attributed to the drive-induced higher-order terms which involve local multi-spin interactions thus give rise to effective short-range interactions between the monopoles. The competition between driveinduced attraction and quantum tunneling may significantly enrich the possible forms of anomalous monopole dynamics. p. 4

Where the paper says it

In that case, quantum tunneling will make it easier for the monopoles to escape from the attractive potential, thereby reducing their trapping time. The competition between driveinduced attraction and quantum tunneling may significantly enrich the possible forms of anomalous monopole dynamics.

Quantum interference between different monopole paths might qualitatively alter their long-time dynamics and lead to anomalous diffusion. p. 4
Where the paper says it

The competition between driveinduced attraction and quantum tunneling may significantly enrich the possible forms of anomalous monopole dynamics. Moreover, quantum interference between different monopole paths might qualitatively alter their long-time dynamics and lead to anomalous diffusion.

The situation with a finite density of monopoles may be even more interesting as the emergent attractive interactions could lead to unprecedented collective behaviors absent in conventional spin-ice systems. p. 4
Where the paper says it

These emergent properties appear only when heating is well suppressed, and no dynamical frustration is observed under slow driving (see details in [38]). We focus only on the dynamics of a pair of monopoles, while the situation with a finite density of monopoles may be even more interesting, as the emergent attractive interactions could lead to unprecedented collective behaviors absent in conventional spin-ice systems.

REFERENCES

arXiv:2511.09401v1[cond-mat. p. 0

Where the paper says it

The Hamiltonian H(t) within one period T reads: H(t) =    J P ⟨ij⟩ S z i S z j ; nT < t ≤ (n + 1 3 )T g P i S x i ; (n + 1 3 )T < t ≤ (n + 2 3 )T J P ⟨⟨ik⟩⟩ S z i S z k ; (n + 2 3 )T < t ≤ (n + 1)T (1) arXiv:2511. 09401v1 [cond-mat.

stat-mech] 3,five or more words copied word for word from that page

--- Page 3 ---

The equation of motion for spin Si reads: S˙i = hi(t) × Si (2),where n is the number of driving circles and the spin Si = [S x i, Sy i, Sz i] is a classical unit vector (Si = 1) on site i. where ⊠ indicates the grey plaquettes in Fig. p. 1

Where the paper says it

(3) where ω = 2π T is the driving frequency. The zero-order approximation H(0) is a time-averaged Hamiltonian over a period: H(0) = 1 T R T 0 dtH(t), which takes the form: 3H(0) = J X ⊠ X ij∈⊠ S z i S z j + g X i S x i , (4) where ⊠ indicates the grey plaquettes in Fig.

1 (b). H(0) describes a 2D counterpart of SI model

--- Page 4 ---

To monitor the monopole trajectories, we define δSz i = S z i(tn)−S˜z i(tn) to measure the difference between the two spin states at t = tn starting from the initial states with and without the initial spin flip Since the movement of a monopole is accompanied by flipping a sequence of spins with alternating S z i, we further define δSz n = P i δSz i(tn) as the total number of the flipped spins during the movement to measure the length of the monopole trajectories Crucially, these long-lived plateaus occur only when two monopoles collide and stick to each other. To see this, we calculate the SI parameter in each plaquette: D⊠ n = P i∈⊠ S z i(tn) (D⊠ n ≈ 2 indicates a monopole located in plaquette ⊠) to locate the position of a monopole[38]

--- Page 5 ---

As detailed in Ref. [38], we find that heating is typically initiated by the collision of two monopoles Instead of being perfectly localized quasiparticles as in the fast-driven case, the two monopoles here smear out and remain bound together for a long time. p. 4

Where the paper says it

[38], we find that heating is typically initiated by the collision of two monopoles. Instead of being perfectly localized quasiparticles as in the fast-driven case, the two monopoles here smear out and remain bound together for a long time.

They generate a region of high energy density, thereby opening new heating channels that enable rapid energy absorption. p. 4
Where the paper says it

Instead of being perfectly localized quasiparticles as in the fast-driven case, the two monopoles here smear out and remain bound together for a long time. They generate a region of high energy density, thereby opening new heating channels that enable rapid energy absorption.

Before heating begins, the two monopoles wander through the lattice, and the typical timescale for their collision during this 2D random walk scales as L2. p. 4
Where the paper says it

Hence, heating spreads from a “seed” (the location of the collision) to the entire lattice, and the ice rule is also broken accordingly. Before heating begins, the two monopoles wander through the lattice, and the typical timescale for their collision during this 2D random walk scales as L 2 , which explains the finite-size dependence of P(τ0).

--- Page 6 ---

In summary, we propose a dynamical frustration mechanism in periodically driven systems, which exhibit a long-tailed, power-law distribution of the waiting time for monopole random walks Drive-induced processes play a key role in monopole dynamics under fast and intermediate driving: they lead to a short-range attraction between monopoles and trap them in the former. These emergent properties appear only when heating is well suppressed, and no dynamical frustration is observed under slow driving (see details in [38]). p. 4

Where the paper says it

Drive-induced processes play a key role in monopole dynamics under fast and intermediate driving: they lead to a short-range attraction between monopoles and trap them in the former, while initiating heating at late times in the latter. These emergent properties appear only when heating is well suppressed, and no dynamical frustration is observed under slow driving (see details in [38]).

We focus only on the dynamics of a pair of monopoles while the situation with a finite density of monopoles may be even more interesting as the emergent attractive interactions could lead to unprecedented collective behaviors absent in conventional spin-ice systems. p. 4
Where the paper says it

These emergent properties appear only when heating is well suppressed, and no dynamical frustration is observed under slow driving (see details in [38]). We focus only on the dynamics of a pair of monopoles, while the situation with a finite density of monopoles may be even more interesting, as the emergent attractive interactions could lead to unprecedented collective behaviors absent in conventional spin-ice systems.

--- Page 7 ---

We expect that similar attractive interactions will also appear in quantum systems and a natural question is whether the long-tail dynamics still persists In that case, quantum tunneling will make it easier for the monopoles to escape from the attractive potential thereby reducing their trapping time The competition between driveinduced attraction and quantum tunneling may significantly enrich the possible forms of anomalous monopole dynamics. Quantum interference between different monopole paths might qualitatively alter their long-time dynamics and lead to anomalous diffusion

--- Page 8 ---

Acknowledgments.—We thank Roderich Moessner and Andrea Pizzi for stimulating discussions ZC is supported by the National Key Research and Development Program of China (2024YFA1408303,2020YFA0309000), Natural Science Foundation of China (Grant No. p. 4

Where the paper says it

— We thank Roderich Moessner and Andrea Pizzi for stimulating discussions. ZC is supported by the National Key Research and Development Program of China (2024YFA1408303,2020YFA0309000), Natural Science Foundation of China (Grant No.

12174251 and No. p. 4
Where the paper says it

ZC is supported by the National Key Research and Development Program of China (2024YFA1408303,2020YFA0309000), Natural Science Foundation of China (Grant No. 12174251 and No.

1252547), Shanghai Municipal Science and Technology Major Project (Grant No. p. 4
Where the paper says it

12174251 and No. 12525407), Shanghai Municipal Science and Technology Major Project (Grant No.

26Z51) ZC is supported by the National Key Research and Development Program of China (No. p. 4
Where the paper says it

— We thank Roderich Moessner and Andrea Pizzi for stimulating discussions. ZC is supported by the National Key Research and Development Program of China (2024YFA1408303,2020YFA0309000), Natural Science Foundation of China (Grant No.

2024ZD0301800) and by the National Natural Science Foundation of China (Grant No. p. 4
Where the paper says it

HZ is supported by Quantum Science and Technology-National Science and Technology Major Project (No. 2024ZD0301800) and by the National Natural Science Foundation of China (Grant No.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI system can predict non-ergodic stochastic random walk processes in frustrated magnetic systems by incorporating drive-induced short-range attractive interactions between emergent monopole excitations.

  2. The system can model long-tailed power-law distributed waiting times for monopole dynamics, where the exponent is tunable by the effective temperature Teff, which can be mapped to the initial noise strength W.

  3. The AI system can forecast heating dynamics in intermediate driving regimes by predicting the duration of prethermal plateaus (tau0), leveraging finite-size scaling results showing that for large systems, algebraic decay is replaced by an exponential decay dependent on system size L.

  4. The improved system can understand the physical origin of long-tail distributions in heating time, identifying that the energy barrier V induced by their attractive force leads to a distribution p(tau) tau-eta where eta = 1 + alphaTeff.

Abstract

In this study, we show that dynamical frustration can spontaneously emerge in frustration-free magnetic systems under periodic driving. Specifically, we consider a classical spin system and demonstrate the emergence of spin-ice physics when drive-induced heating is well suppressed. In particular, we focus on the dynamics of magnetic monopole excitations, which, in sharp contrast to their equilibrium counterparts, exhibit a non-ergodic stochastic random-walk process with long-tailed, power-law distributed waiting times, where the power-law exponent is tunable by the system's effective temperature. Heating is accelerated at intermediate driving frequencies, and the system eventually heats up to an infinite-temperature state. However, the heating time is extremely sensitive to different initial-state realizations and also follows a long-tailed power-law distribution. We show that a drive-induced short-range attractive interaction between monopoles is responsible for the long-tailed distributions observed in both monopole and heating dynamics.

Sources

Related papers