Quantum Hilbert Space Fragmentation and Entangled Frozen States

This paper identifies rank deficiency in local Hamiltonians as the fundamental mechanism driving quantum Hilbert space fragmentation, which generates entangled frozen states that split mobile classical sectors into distinct quantum subspaces exhibiting either weak or strong fragmentation characterized by specific spectral statistics.

The Gist

Imagine the universe of quantum physics as a giant, chaotic dance floor. Usually, when you turn on the music (the Hamiltonian), everyone dances together, mixing and mingling until the whole room is a uniform blur of motion. This is called thermalization or ergodicity—everything eventually explores every possible spot on the floor.

But sometimes, the dance floor gets broken up. Instead of one big room, it gets divided into thousands of tiny, isolated booths. People in one booth can't talk to people in another. This is called Hilbert Space Fragmentation.

For a long time, physicists thought this only happened because of "rules" (symmetries) or "bad luck" (disorder). But this paper discovers a new, sneaky reason why the dance floor breaks: Rank Deficiency.

Here is the story of the paper, explained through a few creative analogies.

1. The Broken Dance Move (The Mechanism)

Imagine a rule for the dance: "If three people stand in a row wearing the same color shirt, they must swap places."

• Classical Fragmentation: In the old models, if you had a line of people, some groups could move around freely, while others were stuck. The "frozen" ones were just people standing still because the rules didn't apply to them.
• The New Discovery: The authors found that sometimes, the "swap" rule is broken in a specific way. It's like a dance instructor who says, "You can swap, but only if you do it in a specific, complicated pattern."
  • If you try to swap in a simple way, the instructor stops you.
  • However, there is a hidden, entangled pattern (a specific combination of moves) where the instructor doesn't stop you, but also doesn't let you move anywhere else. You are stuck in a "frozen" state, but you are entangled (your moves are perfectly synchronized with your neighbors in a way that looks like a single, complex object).

The authors call these Entangled Frozen States (EFS). They are like dancers who are technically "frozen" in place, but they are frozen in a complex, wavy, quantum pose that no single person could hold alone.

2. The Two Types of Broken Floors (Weak vs. Strong)

The paper divides these broken dance floors into two categories, depending on how the "mobile" dancers (the ones who can still move) behave after the frozen ones are removed.

Weak Fragmentation: The VIP Lounge

Imagine the dance floor breaks, but the remaining mobile dancers are all crammed into one or two huge, VIP lounges.

• Inside these lounges, everyone still dances wildly and mixes with everyone else.
• The system is "ergodic" inside the lounge.
• The Analogy: It's like a party where the room is locked, but inside the room, it's a total rave. The chaos is contained, but it's still chaotic.
• The Models: The "Asymmetric" and "GHZ" models in the paper fit here. Even though the floor is broken, the mobile part is still one big, connected block.

Strong Fragmentation: The Infinite Maze

Now imagine the dance floor breaks into millions of tiny, individual cubicles.

• Each dancer is trapped in their own tiny box.
• They can't talk to anyone else.
• The Analogy: It's like a prison with infinite cells. Everyone is alone. The system is completely frozen and disordered.
• The Model: The famous "Temperley-Lieb" model fits here. The rules of the game (mathematical relations called the Jones relation) are so strict that they chop the mobile space into millions of tiny, isolated pieces.

3. The Role of Symmetry (The Bouncer)

You might think you need a "Bouncer" (Symmetry) to break the dance floor.

• The Paper's Verdict: No! You don't need a bouncer. The floor breaks just because the dance move itself is "rank-deficient" (broken in a specific mathematical way).
• What Symmetry Does: If a bouncer is present, they just organize the booths. They might put all the "Red Shirt" dancers in one VIP lounge and "Blue Shirt" dancers in another. They make the structure more orderly, but they didn't cause the breakage in the first place.

4. The "Gap-Ratio" Test (The Music Check)

How do the authors know if the dance floor is "Weak" or "Strong"? They listen to the music (the energy levels of the system).

• Weak Case (The Rave): The music sounds like a complex, chaotic jazz song (GOE statistics). The notes are tightly packed and repel each other, showing that the dancers are interacting.
• Strong Case (The Silence): The music sounds like random, isolated beeps (Poisson statistics). The notes are scattered and don't care about each other, showing that the dancers are trapped in their own boxes.

Summary

This paper solves a mystery: Why do some quantum systems get stuck?

1. The Cause: It's not always about symmetry or bad luck. Sometimes, the local rules of the game are just "underpowered" (rank-deficient), creating hidden, entangled states that act like anchors.
2. The Result: These anchors freeze parts of the system, splitting the dance floor.
3. The Spectrum:
   • Sometimes the remaining dancers are still in one big, chaotic room (Weak Fragmentation).
   • Sometimes the room is shattered into infinite tiny cells (Strong Fragmentation).

The authors proved this using four different "dance models," ranging from simple, asymmetric ones to complex, highly structured ones, showing that this mechanism is a universal rule of quantum mechanics, not just a fluke of one specific system.

Technical Summary

1. Problem Statement

Hilbert Space Fragmentation (HSF) is a mechanism for ergodicity breaking where the Hilbert space decomposes into exponentially many dynamically disconnected Krylov subspaces, distinct from Many-Body Localization (which relies on disorder) or Quantum Many-Body Scars (which rely on fine-tuned eigenstates).

• Classical HSF: Krylov sectors are spanned by product states. This is well-understood via semigroup word problems.
• Quantum HSF: The classical Krylov sectors themselves decompose into sub-sectors spanned by entangled states. This phenomenon is poorly understood.
• Open Questions:
  1. What is the minimal ingredient required for quantum fragmentation? (Is complex algebraic structure like the Temperley-Lieb algebra essential?)
  2. What role do symmetries play? (Are they a prerequisite or an optional add-on?)
  3. How can quantum fragmentation be classified (e.g., weak vs. strong)?

2. Methodology

The authors employ a framework combining semigroup dynamics (classical rewriting rules) and local Hamiltonian analysis (quantum constraints).

• Model Construction: They define a 1D chain with local Hilbert space dimension $d$. The dynamics are governed by a semigroup presentation $\tilde{G} = \langle S | R \rangle$, where $R$ defines equivalence relations between local words.
• Hamiltonian Formulation: The local Hamiltonian terms are constructed as projectors onto specific states within the equivalence classes:
  $$h_i = \sum_{\alpha} \sum_{w', w \in R_\alpha} g_{w', w} |w'\rangle\langle w|$$
  The total Hamiltonian is $H = \sum J_i h_i$.
• Key Mechanism Analysis: The authors investigate the rank deficiency of the coupling matrix $g_{w', w}$ within an equivalence class.
  • If the matrix is full rank, the local term explores all directions, and no quantum fragmentation occurs.
  • If the matrix is rank-deficient ($r < |R_\alpha|$), it creates local "null directions."
• Entangled Frozen States (EFS): The intersection of these local null directions across the system defines a subspace of states that are:
  1. Entangled: They cannot be product states (as product states in mobile sectors evolve).
  2. Frozen: They are annihilated by the Hamiltonian ($H|\psi\rangle = 0$) and do not evolve.
• Decomposition: A mobile classical Krylov sector $K_{cl}$ splits into:
  $$K_{cl} = K_q \oplus K_{EFS}$$
  Where $K_q$ is the mobile quantum Krylov subspace and $K_{EFS}$ is the entangled frozen subspace.
• Classification: The authors introduce Weak vs. Strong Quantum Fragmentation based on the reducibility of $K_q$:
  • Weak: $K_q$ contains at least one irreducible block occupying a finite fraction of the space ($D_{max}/D_q \sim O(1)$).
  • Strong: $K_q$ decomposes into a number of irreducible blocks that grows with system size, such that $D_{max}/D_q \to 0$.

3. Key Contributions

The paper establishes a universal mechanism for quantum fragmentation and classifies it through four models of increasing symmetry and algebraic structure:

1. Identification of the Universal Mechanism:

   • The authors prove that rank deficiency of local coupling terms in a classically fragmented model is the sufficient and minimal condition for quantum fragmentation.
   • Symmetry is not required. Quantum fragmentation exists even in asymmetric models. Symmetries merely organize the resulting subspaces (creating degeneracies or charge sectors) but do not generate the fragmentation itself.

2. Four Model Demonstrations:

   • Asymmetric Triplet-Flip Projector: A qubit chain with $000 \leftrightarrow 111$ rules and an asymmetric projector ($a|000\rangle + b|111\rangle$). This model has no symmetry. It exhibits quantum fragmentation where each mobile sector splits into one EFS and one irreducible mobile quantum block.
   • GHZ Projector ($Z_2$ Symmetry): Restores $Z_2$ spin-flip symmetry ($a=b$). The mechanism remains the same, but the symmetry organizes the mobile blocks into degenerate pairs or splits the all-mobile sector into charge sectors.
   • Cyclic Qutrit Projector ($Z_3$ Symmetry): Uses a cyclic qutrit model with $Z_3$ symmetry. The EFS dimension grows with system size. The symmetry decomposes the mobile space into charge sectors.
   • Temperley-Lieb (TL) Model: Incorporates the Jones relation (algebraic constraint). Here, the mobile space $K_q$ is further decomposed by the TL algebra into many irreducible modules, not just by symmetry.

3. Weak vs. Strong Quantum Fragmentation:

   • Weak Fragmentation: Observed in the Asymmetric, GHZ, and Cyclic models. The mobile quantum space $K_q$ remains largely irreducible (or splits into a constant number of blocks, $O(1)$).
   • Strong Fragmentation: Observed in the TL model. The Jones relation forces $K_q$ to split into an exponentially growing number of irreducible blocks as system size $L \to \infty$.

4. Spectral Signatures:

   • The authors link the fragmentation type to gap-ratio statistics ($r$).
   • Weak Case: The spectrum follows an $m$-GOE distribution (superposition of $m$ independent Gaussian Orthogonal Ensemble distributions), where $m$ is the number of irreducible blocks (e.g., $m=1$ for asymmetric, $m=2$ for GHZ, $m=3$ for cyclic).
   • Strong Case: As the number of irreducible blocks $N_{irr} \to \infty$, the gap-ratio distribution converges to Poisson, indicating a lack of level repulsion typical of integrable or localized systems.

4. Results

• Closed-Form Expressions: For the asymmetric and GHZ models, the authors derived exact formulas for Krylov dimensions, degeneracies, and sector multiplicities.
• EFS Construction: They provided an explicit inductive construction (Ind map) for Entangled Frozen States, showing how local constraints propagate to create global entangled states that are dynamically frozen.
• Numerical Verification:
  • Simulations up to $L=18$ confirm the theoretical predictions.
  • Figure 3 demonstrates the transition in gap-ratio statistics:
    • Asymmetric ($L=18$): Drifts toward GOE ($m=1$).
    • GHZ ($L=18$): Matches 2GOE ($m=2$).
    • Cyclic ($L=12$): Matches 3GOE ($m=3$).
    • TL ($L=18$): Approaches Poisson distribution due to the proliferation of TL blocks.

5. Significance

• Theoretical Unification: The paper resolves the ambiguity regarding the origins of quantum fragmentation, showing it is a generic consequence of rank-deficient local constraints in fragmented classical systems, independent of specific algebraic structures like the Temperley-Lieb algebra.
• New Classification: The introduction of "Weak" vs. "Strong" quantum fragmentation provides a rigorous framework to categorize ergodicity breaking beyond the classical dichotomy.
• Experimental Relevance: The GHZ projector model is identified as a candidate for efficient Trotterization on near-term quantum hardware, offering a concrete platform to observe the weak/strong transition and probe entangled frozen states.
• Connection to Error Correction: The existence of EFS, which are dynamically protected by entanglement structure rather than symmetry, suggests potential connections to quantum error correction codes.

In summary, this work establishes that rank deficiency is the fundamental engine of quantum Hilbert space fragmentation, generating Entangled Frozen States that split mobile classical sectors. It further classifies these systems based on the reducibility of the remaining mobile quantum space, distinguishing between weak fragmentation (GOE-like statistics) and strong fragmentation (Poisson-like statistics).