Frozen Motion: How Interacting Particles Can Halt in Time

Author: Denis Avetisyan

New research reveals a surprising state of matter where many-body interactions prevent particles from exploring all possible states, leading to a form of ‘dynamical localization’.

The study demonstrates how a driven, two-mode bosonic system, modeled within a Fock space framework and subject to two-body interactions, exhibits a transition from quantum to classical evolution-a shift evidenced by the divergence of mean-field trajectories and Husimi distributions at increasing periods, beginning with initial states <span class="katex-eq" data-katex-display="false">\ket{\psi\_{0}}=\ket{100,100}</span> and characterized by parameters such as rotation angle <span class="katex-eq" data-katex-display="false">a=0.1</span> and torsion amplitude <span class="katex-eq" data-katex-display="false">b=600</span>. — The study demonstrates how a driven, two-mode bosonic system, modeled within a Fock space framework and subject to two-body interactions, exhibits a transition from quantum to classical evolution-a shift evidenced by the divergence of mean-field trajectories and Husimi distributions at increasing periods, beginning with initial states $\ket{\psi\_{0}}=\ket{100,100}$ and characterized by parameters such as rotation angle $a=0.1$ and torsion amplitude $b=600$ .

This study demonstrates many-body dynamical localization in a kicked Bose-Hubbard model, revealing connections to discrete time crystals and the breakdown of ergodicity in quantum systems.

The breakdown of ergodicity in strongly interacting quantum systems remains a fundamental challenge in many-body physics. In this work, ‘Many-body dynamical localization in Fock space’, we investigate the emergence of dynamical localization in a minimal model of periodically driven, interacting bosons, revealing a surprising suppression of transport in the system’s Fock space. We demonstrate that this many-body dynamical localization-characterized by a localization length scaling with system parameters and a crossover to random-matrix statistics-exhibits connections to both Anderson localization and the physics of discrete time crystals. Could this framework provide a novel route to understanding and controlling quantum transport in complex systems?

The Illusion of Order: Beyond Static Localization

For decades, the prevailing understanding of quantum localization centered on the role of static disorder – imperfections or irregularities fixed in space that trap quantum particles. However, this perspective fails to capture the behavior of many physical systems where the disruptive influence isn’t constant, but rather fluctuates over time. Real-world environments, such as materials subjected to oscillating fields or quantum systems interacting with dynamic backgrounds, introduce time-dependent perturbations that profoundly alter particle behavior. These temporal fluctuations can induce localization even in systems lacking spatial disorder, creating a fundamentally different mechanism where motion is hindered not by what is there, but by how things change. Consequently, a complete picture of localization requires moving beyond static landscapes and embracing the complexities of time-driven dynamics.

The conventional understanding of localization – the trapping of quantum particles – often relies on spatial disorder, such as imperfections in a material’s structure. However, this framework falls short when applied to systems subject to time-varying forces or periodic modulations. Recent investigations demonstrate that localization can, in fact, emerge as an inherent property of the dynamics of a system, even in a perfectly ordered lattice. This time-driven localization arises not from where something is, but from how it moves, with carefully engineered temporal changes creating effective barriers to particle propagation. Such mechanisms offer a pathway to control localization without physically altering the system’s structure, potentially enabling novel applications in areas like wave manipulation and quantum information processing, and broadening the scope of phenomena considered within the field of many-body localization.

The potential to manipulate quantum localization through temporal dynamics represents a significant leap beyond traditional approaches. While static disorder has long been considered the primary driver of localized states, emerging research demonstrates that carefully engineered time-dependent perturbations can induce and control these effects. This offers the tantalizing prospect of designing systems where localization isn’t a passive consequence of imperfections, but an actively tunable property. Such control could unlock applications in areas like quantum information processing, where localized states are crucial for maintaining qubit coherence, and in novel material design, enabling the creation of systems with tailored electronic and transport properties. The ability to ‘switch’ localization on and off, or to dynamically reshape localized states, promises a new era of quantum control and a deeper understanding of non-equilibrium quantum phenomena.

Simulations demonstrate that population imbalance variance <span class="katex-eq" data-katex-display="false">\sigma_{z}^{2}(m)</span> exhibits both diffusive and localized behaviors depending on modulation strength <span class="katex-eq" data-katex-display="false">a</span>, with exponential saturation observed for localization and scaling with atom number <span class="katex-eq" data-katex-display="false">N</span> as predicted by equations (5) and (7). — Simulations demonstrate that population imbalance variance $\sigma_{z}^{2}(m)$ exhibits both diffusive and localized behaviors depending on modulation strength $a$ , with exponential saturation observed for localization and scaling with atom number $N$ as predicted by equations (5) and (7).

The Quantum Dance: Modeling Periodic Perturbations

The quantum kicked top serves as a well-defined, controllable model system for investigating the dynamics of quantum systems subject to periodic perturbations. This system analogizes a particle experiencing discrete impulses at regular time intervals, allowing researchers to study the effects of these impulses on quantum state evolution. Unlike continuous driving, the kicked top’s impulsive nature simplifies the mathematical treatment while retaining key features of periodically driven systems, such as the emergence of phenomena like dynamical localization and chaotic behavior. The model’s parameters, including the kick strength and period, can be adjusted to explore a wide range of driving regimes and their impact on quantum dynamics, providing insights applicable to diverse physical scenarios like atomic physics and condensed matter systems.

The Floquet operator, denoted as $\hat{U}$ , is a unitary operator that describes the time evolution of a quantum system subjected to periodic driving. For the quantum kicked top, this operator encapsulates the combined effect of free evolution between kicks and the instantaneous kick itself. Mathematically, $\hat{U} = \hat{T} \exp \left( -i \in t_0^T \hat{H}(t) dt \right)$ , where $\hat{H}(t)$ is the time-dependent Hamiltonian and $\hat{T}$ denotes time ordering. Analyzing the spectrum of the Floquet operator – specifically its eigenvalues and eigenvectors – allows for the determination of quasi-energies and quasi-momentum states, effectively transforming the time-dependent problem into an equivalent time-independent one. This approach provides a rigorous mathematical framework for understanding the long-term dynamics and stability of the system under periodic perturbations, enabling precise calculations of observable quantities.

The quantum kicked top utilizes Hilbert space, and specifically Fock space, to mathematically define the possible states of the system and track their temporal evolution. Fock space is a state space that describes multi-particle quantum systems; in this model, it accounts for any number of bosons that may occupy a given energy level. Each basis state within the Fock space corresponds to a specific number of bosons, and operators acting on these states describe changes in particle number and energy. The use of Fock space allows for a complete and consistent description of the quantum kicked top’s dynamics, enabling the calculation of probabilities for transitions between different quantum states following periodic driving.

The quantum kicked top model utilizes bosons as the constituent particles defining the system’s quantum states. These particles, possessing integer spin, obey Bose-Einstein statistics, allowing multiple bosons to occupy the same quantum state simultaneously. Within the model, bosonic operators $a$ and $a^{\dagger}$ are employed to describe the creation and annihilation of these particles, forming the basis for constructing the system’s Hamiltonian and analyzing its dynamics under periodic driving. The use of bosons simplifies the mathematical treatment compared to fermionic systems and is crucial for accurately representing the collective behavior of the top under repeated impulses.

Analysis using the many-body-delocalized-localization framework reveals that discrete time crystals exhibit population imbalances dependent on driving phase <span class="katex-eq" data-katex-display="false">m</span>, degeneracy lifting proportional to <span class="katex-eq" data-katex-display="false">|\Delta\varepsilon\_{\pm}-h/2T|</span>, and a localization crossover scaling with both driving phase and system size <span class="katex-eq" data-katex-display="false">N</span>, as evidenced by the behavior of <span class="katex-eq" data-katex-display="false">\sigma\_{z}^{2}</span>. — Analysis using the many-body-delocalized-localization framework reveals that discrete time crystals exhibit population imbalances dependent on driving phase $m$ , degeneracy lifting proportional to $|\Delta\varepsilon\_{\pm}-h/2T|$ , and a localization crossover scaling with both driving phase and system size $N$ , as evidenced by the behavior of $\sigma\_{z}^{2}$ .

The Illusion of Simplicity: Reducing Complexity with Effective Hamiltonians

The quantum kicked top, a paradigmatic model in chaotic quantum mechanics, possesses a complex phase space described by angular momentum variables. An effective Hamiltonian facilitates analysis by isolating the degrees of freedom most pertinent to the system’s dynamics; typically, this involves focusing on the action variable $J_z$ and the angle variable θ. This reduction in complexity significantly streamlines calculations of time evolution and spectral properties. By representing the system with a lower-dimensional Hamiltonian, researchers can more readily identify key features such as the emergence of cantori and the transition to chaos, and investigate phenomena like dynamical localization without being overwhelmed by the full Hilbert space dimensionality. The effective Hamiltonian approach enables the extraction of analytical and numerical results that would otherwise be computationally intractable.

The construction of an effective Hamiltonian for the quantum kicked top necessitates the use of angular momentum coupling theory, specifically employing the Wigner d-matrix $D_{m_1 m_2 m_3 m_4}$ . This matrix element facilitates the transformation between different coupling schemes of three angular momenta, allowing for the decoupling of irrelevant degrees of freedom and the simplification of the Hilbert space. By representing angular momentum states in a basis where the total angular momentum is well-defined, the Wigner d-matrix enables the systematic reduction of the system’s complexity. This transformation is crucial for isolating the dynamics relevant to the kicking potential and for efficiently calculating matrix elements within the reduced Hilbert space, ultimately leading to a tractable Hamiltonian that captures the essential physics of the system.

Dynamical localization in the quantum kicked top, achieved through periodic application of a δ-kick potential, represents a phenomenon where the wave packet’s spreading is suppressed, confining the particle to a limited region of phase space. This localization occurs despite the complete absence of static disorder or imperfections in the system, distinguishing it from Anderson localization. The periodic kicking effectively creates a quasi-crystal potential in phase space, leading to the formation of resonances and the suppression of diffusion. Specifically, the kicking introduces correlations between successive iterations, hindering the particle’s ability to explore all available states and resulting in a localized regime for sufficiently strong kicking strength. This is demonstrated through analysis of the Floquet operator and its eigenstates, which reveal a fractal structure in phase space indicative of localization.

Simulations reveal that population imbalance variance <span class="katex-eq" data-katex-display="false">\sigma_{z}^{2}(m)</span> evolves differently for classical and quantum systems depending on initial conditions-near the poles or equator-and that disorder-averaged Fock-space projections demonstrate localization behavior characterized by a diffusion coefficient <span class="katex-eq" data-katex-display="false">\xi = D/\hbar_{eff}^{2}</span>, although negative diffusion can limit its effectiveness in certain quantum states. — Simulations reveal that population imbalance variance $\sigma_{z}^{2}(m)$ evolves differently for classical and quantum systems depending on initial conditions-near the poles or equator-and that disorder-averaged Fock-space projections demonstrate localization behavior characterized by a diffusion coefficient $\xi = D/\hbar_{eff}^{2}$ , although negative diffusion can limit its effectiveness in certain quantum states.

Beyond Randomness: Quantifying the Signature of Localization

Many physical systems exhibit localization, where particles or excitations become confined to specific regions of space. Describing the statistical fluctuations within these localized systems proves challenging because traditional probability distributions, such as the Poisson distribution, are fundamentally built on assumptions of spatial homogeneity and independence – assumptions routinely violated when dealing with constrained environments. The Poisson distribution predicts that the variance of a measured quantity will equal its mean, a relationship that holds true for uncorrelated events spread evenly throughout a system. However, in localized scenarios, correlations emerge as entities interact within the confined space, and the very act of localization reduces the effective number of independent events. Consequently, observed fluctuations often deviate substantially from Poissonian behavior, revealing a lower variance and signaling the need for more sophisticated statistical tools capable of capturing the unique characteristics of these constrained systems.

Quantifying deviations from expected randomness requires a sensitive metric, and the Kullback-Leibler divergence (DKL) provides just that-a precise measure of the difference between an observed probability distribution and a standard Poisson distribution. Unlike simple visual inspection, DKL assigns a numerical value representing the ‘information gain’ achieved by using the observed distribution instead of the Poisson prediction; a larger DKL indicates a stronger departure from randomness. Crucially, this divergence isn’t merely descriptive, but allows for rigorous comparison across different systems and parameter regimes. By calculating $D_{KL}$ , researchers can objectively assess the degree to which a system exhibits non-Poissonian behavior, offering a quantitative fingerprint of phenomena like dynamical localization and providing a pathway to link observable statistical deviations to underlying physical mechanisms and system properties.

Investigations into dynamical localization reveal a departure from the predictions of standard Poisson statistics, a finding substantiated through the application of Kullback-Leibler divergence $D_{KL}$ . This measure quantifies the dissimilarity between observed probability distributions and a purely Poissonian expectation, and its application to localized systems consistently demonstrates a significant deviation. Specifically, the observed variance in population imbalance falls below the $1/3$ limit predicted by Poisson statistics, signaling a fundamentally altered behavior. This confirms that dynamical localization isn’t merely a reduction in movement, but a genuine reshaping of probabilistic outcomes, validating the unique characteristics of this quantum phenomenon and providing a pathway to characterize its strength based on measurable deviations from established statistical norms.

The degree to which a system exhibits dynamical localization isn’t merely a qualitative observation, but a quantifiable phenomenon directly linked to controllable system parameters. Analysis reveals that deviations from a standard Poisson distribution-measured through metrics like Kullback-Leibler divergence-correlate with the localization length, $ξ ≈ D/ℏ_{eff}^2$ , where $D$ represents the disorder strength and $ℏ_{eff}$ the effective Planck constant. Critically, in strongly localized regimes, the variance of population imbalance-a measure of particle distribution-decreases below the classical limit of 1/3, indicating a suppression of fluctuations. This connection allows for predictive control; by tuning parameters like disorder strength, researchers can actively manipulate the localization length and, consequently, the system’s behavior, opening avenues for designing materials with tailored quantum properties and enhanced stability against decoherence.

Spectral statistics of the RQKT model demonstrate a crossover from random matrix theory (RMT) behavior to localization as the parameter <span class="katex-eq" data-katex-display="false">a</span> increases, confirmed by the Kullback-Leibler divergence <span class="katex-eq" data-katex-display="false">D_{KL}</span> which aligns with the theoretically predicted localization crossover at <span class="katex-eq" data-katex-display="false">N = 16 / \sin^2(a)</span>. — Spectral statistics of the RQKT model demonstrate a crossover from random matrix theory (RMT) behavior to localization as the parameter $a$ increases, confirmed by the Kullback-Leibler divergence $D_{KL}$ which aligns with the theoretically predicted localization crossover at $N = 16 / \sin^2(a)$ .

The study of many-body dynamical localization presents a humbling experience for theoretical physics. It reveals how even meticulously constructed models, designed to capture quantum behavior, can exhibit unforeseen limitations. This mirrors the inevitable boundary faced by any attempt to fully grasp reality. As Aristotle observed, “The ultimate value of life depends upon awareness and the power of contemplation rather than upon mere survival.” The breakdown of ergodicity, highlighted within the paper’s exploration of interacting bosons, serves as a potent reminder: any theory is good until light leaves its boundaries. The system’s transition into a localized state demonstrates that the pursuit of complete predictability is, ultimately, an exercise in defining the limits of knowledge itself.

Where Do We Go From Here?

The demonstration of many-body dynamical localization within a periodically kicked Bose system, and its apparent kinship with discrete time crystals, serves less as a resolution and more as a sharpening of existing paradoxes. The Floquet operator, a mathematical convenience allowing treatment of time-periodic systems, ultimately obscures the question of whether ergodicity truly fails, or merely retreats beyond the limits of practical observation. Any insistence on a precise definition of ‘localization’ in Fock space invites scrutiny; the Hilbert space itself is a construct, and the interpretation of any observable – a measurable quantity – is always contingent.

Future investigations must confront the limitations inherent in minimal models. While the quantum kicked rotor offers analytical tractability, its distance from realistic many-body systems is considerable. The introduction of longer-range interactions, or disorder beyond the simple potential considered here, may reveal subtle instabilities or unexpected emergent behaviors. Furthermore, a rigorous examination of finite-size effects is crucial; the apparent localization observed may simply be a consequence of limited system size, vanishing as the number of bosons increases.

The pursuit of ‘breakdown of ergodicity’ may ultimately prove a fool’s errand. The universe does not require justification of its non-ergodic tendencies. Instead, perhaps the value lies in recognizing that any attempt to impose order-any theoretical framework-is itself subject to the same fundamental limitations. The event horizon of complexity is always approaching.

Original article: https://arxiv.org/pdf/2604.09224.pdf

Contact the author: https://www.linkedin.com/in/avetisyan/

The Illusion of Order: Beyond Static Localization

The Quantum Dance: Modeling Periodic Perturbations

The Illusion of Simplicity: Reducing Complexity with Effective Hamiltonians

Beyond Randomness: Quantifying the Signature of Localization

Where Do We Go From Here?

See also: