Atypical Dynamics & Many-body Scars

Quantum many-body scars describe structured exceptions to otherwise thermalizing dynamics. Certain initial states can exhibit unusually persistent revivals, and selected eigenstates can have atypical entanglement or matrix elements. Rozon and Agarwal's contribution is to ask whether the Hamiltonian itself simplifies when restricted to the scar subspace. This perspective links coherent trajectories to operator algebra, while retaining the crucial distinction between exceptional states and generic behavior throughout Hilbert space.

Figure to read

Restricted invariant subspaces distinguish embedding conditions from the broader broken-unitary construction. Figure 2 in the source paper. Rozon, Pierre-Gabriel; Agarwal, Kartiek. Broken unitary picture of dynamics in quantum many-body scars, Physical Review Research 6, 023041 (2024). Paper. CC BY 4.0. Original manuscript graphic; no alterations.

Compare the blocks on which the operator conditions are imposed. The figure illustrates a relaxation of algebraic requirements outside the selected subspace; it is not a measured revival curve.

Why the Rydberg result was surprising

Programmable Rydberg chains provide constrained many-body dynamics: strong interactions suppress nearby simultaneous excitations. Bernien and collaborators observed strikingly different relaxation from different initial configurations in a 51-atom system, providing the experimental context for subsequent theoretical work. The phenomenon was surprising precisely because the constrained model was not simply an integrable system with an obvious complete set of conservation laws. 1

Turner and collaborators connected these dynamics to atypical eigenstates in what became a central example of quantum many-body scarring. In this setting, special initial states overlap strongly with an approximately organized set of eigenstates, allowing coherent returns that generic states do not exhibit. Imperfect energy spacing and coupling to other states limit revival quality. The scar interpretation therefore concerns a structured part of the spectrum and its accessible dynamics, not universal immunity to thermalization. 2

The broken-unitary perspective

Rozon and Agarwal consider decomposing a Hamiltonian as $H=\sum_a O_a$. Although the terms generally fail to commute on the full Hilbert space, they can obey much stronger relations within a special subspace. There, evolution can admit a factorized or “broken-unitary” description built from the constituent evolutions. The distinction between global and restricted commutation is the core idea. Checking an ordinary commutator on the entire space would miss it. 3

This viewpoint is useful because it turns the search for unusual eigenstates into a question about compatible operator actions. It can explain existing scar constructions and suggest ways to build or identify others. The claim must still be applied to the particular models and subspaces analyzed. Approximate scars, exact embedded states, and nearly periodic trajectories are related but not identical mathematical objects; a single slogan should not erase those differences.

The paper distinguishes constructions with a finite number of constituent operators, including examples associated with the AKLT model, from those with an extensive number, including Hubbard-model eta-pairing states. This distinction helps organize which local conditions generate the exceptional subspace and how the construction changes with system size. It gives the algebraic picture concrete model content beyond a general statement about revivals. 3

Complementary construction strategies

Shiraishi and Mori provide a systematic embedding approach to constructing counterexamples to eigenstate thermalization. Their construction demonstrates that selected nonthermal eigenstates can be embedded without making the entire Hamiltonian conventionally integrable. That is complementary to a dynamical analysis: proving an exceptional eigenstate exists does not automatically show that a simple experimentally prepared state will display large, useful revivals. 4

The ETH framework clarifies the distinction between a strong statement about every eigenstate and statements about overwhelmingly typical states. A sparse exceptional set can challenge the former while leaving ordinary thermal behavior intact for generic preparations. This is why a scar experiment should report initial-state selectivity and overlap with the relevant subspace, rather than only a spectacular trajectory. 5

What would make scars useful?

For quantum control, the attraction is coherent many-body dynamics without freezing the whole system. But usefulness requires robustness to preparation errors, perturbations, and coupling to unwanted degrees of freedom. A revival visible for one ideal product state may deteriorate rapidly under small defects. The relevant benchmark is therefore a neighborhood of physically preparable states and realistic Hamiltonians, not solely an exactly chosen vector.

The natural connections are to thermalization diagnostics and symmetry engineering. Both ask which restricted structures can survive for long enough to be useful. The critical-cooling pages address a different target—low-energy preparation—and should not classify moving-front simulations as scar results merely because both involve nonequilibrium evolution. An open research direction is whether restricted algebraic structure can guide robust control designs while maintaining tractable readout. That would connect the explanatory success of scar theory to a measurable advantage under experimental imperfections.

References

1. Bernien, Hannes; Schwartz, Sylvain; Keesling, Alexander; Levine, Harry; Omran, Ahmed; Pichler, Hannes; Choi, Soonwon; Zibrov, Alexander S.; Endres, Manuel; Greiner, Markus; Vuletić, Vladan; Lukin, Mikhail D.. Probing many-body dynamics on a 51-atom quantum simulator. Nature 551, 579-584 (2017). Open manuscript.
2. Turner, Christopher J.; Michailidis, Alexios A.; Abanin, Dmitry A.; Serbyn, Maksym; Papic, Zlatko. Weak ergodicity breaking from quantum many-body scars. Nature Physics 14, 745–749 (2018). Open manuscript.
3. Rozon, Pierre-Gabriel; Agarwal, Kartiek. Broken unitary picture of dynamics in quantum many-body scars. Physical Review Research 6, 023041 (2024). Open manuscript.
4. Shiraishi, Naoto; Mori, Takashi. Systematic Construction of Counterexamples to the Eigenstate Thermalization Hypothesis. Phys. Rev. Lett. 119, 030601 (2017). Open manuscript.
5. D'Alessio, Luca; Kafri, Yariv; Polkovnikov, Anatoli; Rigol, Marcos. From Quantum Chaos and Eigenstate Thermalization to Statistical Mechanics and Thermodynamics. Adv. Phys. 65, 239 (2016). Open manuscript.

Independent research synthesis. Literature checked 5 September 2026; preprints are identified in the references.