⚛️ quantum physics

Decoupling band topology from criticality in bosonic systems

This paper demonstrates that in quadratic bosonic systems, band topology and bulk-boundary correspondence remain robust and independent of dynamical stability, revealing that topological phase transitions are driven by Krein collisions rather than the thermodynamic criticality typically associated with long-range correlations.

Original authors: Mariam Ughrelidze, Lorenza Viola, Emilio Cobanera

Published 2026-07-03

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Mariam Ughrelidze, Lorenza Viola, Emilio Cobanera

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are a conductor leading an orchestra of tiny, invisible particles called bosons. In the world of physics, we usually try to organize these particles into neat patterns, much like arranging musicians in a symphony. One of the most famous ways to organize them is called the "Tenfold Way," a rulebook that works perfectly for fermions (a different type of particle, like electrons).

For fermions, the rulebook says: "If you change the pattern of the music (the topology) in a way that creates a new kind of song, the orchestra must hit a moment of chaos (criticality) where the sound becomes long and echoing before the change happens."

This paper asks a simple question: Does this same rulebook work for bosons?

The authors, Mariam Ughrelidze, Lorenza Viola, and Emilio Cobanera, say: No, it doesn't. They discovered that for bosons, you can completely separate the "pattern change" from the "moment of chaos."

Here is a breakdown of their discovery using simple analogies:

1. The Two Types of "Chaos"

In the bosonic world, there are two different ways an orchestra can get into trouble:

The "Dynamical Instability" (The Melting Pot): Imagine the musicians start playing so wildly that the volume grows forever until the instruments break. This is dynamical instability. The paper shows that this happens at a specific line in their "control room" (parameter space). When the system crosses this line, the sound becomes long and echoing (long-range correlations). This is the "chaos" we usually expect.
The "Topological Transition" (The Pattern Shift): Imagine the musicians suddenly switch from playing a waltz to a tango. The music changes its fundamental shape, but the volume stays controlled. This is a topological transition.

2. The Great Decoupling

In the world of fermions (electrons), you cannot switch from a waltz to a tango without the music first becoming chaotic and echoing. The pattern change forces the chaos.

But for bosons, the authors found something surprising:
You can switch the pattern (the topology) from a waltz to a tango without the music ever becoming chaotic or echoing.

The Analogy: Think of a light switch. For fermions, flipping the switch (changing the pattern) causes a massive power surge (chaos) that lights up the whole room. For bosons, the authors found a special switch where you can flip it, and the light changes color (the pattern changes), but the room stays perfectly quiet and stable. The "chaos" (long-range correlations) only happens if you flip a different switch entirely.

3. The "Ghost" of the Pattern

The authors built a specific model (a "bosonic Su-Schrieffer-Heeger model") to prove this. They found that:

When they changed the "pattern" parameter, the edge of their system (the boundary) developed special "ghost" notes (zero-energy modes) that stuck to the walls.
However, the "echo" in the middle of the room (the bulk correlations) didn't care about this change at all. The echo only got loud when they changed a different parameter that made the system unstable.

4. Why This Matters (According to the Paper)

The paper argues that for bosons, the "topology" (the shape of the pattern) and "criticality" (the moment of chaos) are decoupled. They are independent features.

For Fermions: Topology and Chaos are best friends; they always show up together.
For Bosons: Topology and Chaos are strangers. You can have a topological change with no chaos, or chaos with no topological change.

The authors conclude that the "rules" we learned from fermions (like electrons in a metal) do not apply directly to bosons. In the bosonic world, you can have a topological phase transition that looks perfectly calm and stable, even though the pattern has completely changed. The "chaos" is not a required guest at the party; it only shows up if you invite it via a different door (dynamical instability).

In short: The paper reveals that in the bosonic universe, you can change the fundamental shape of reality without causing a meltdown. The two concepts, which are tightly linked in the electron world, are completely independent in the boson world.

Technical Summary: Decoupling Band Topology from Criticality in Bosonic Systems

Problem Statement
In fermionic systems, the "tenfold way" classification establishes a tight link between bulk topological phase transitions, boundary physics (bulk-boundary correspondence), and quantum criticality. Specifically, for free fermions, a change in the topological invariant of a gapped Hamiltonian necessitates the closing of the energy gap, which simultaneously drives the divergence of the correlation length in the ground state (criticality).

This paper investigates whether this tight coupling holds for quadratic bosonic Hamiltonians (QBHs). The authors highlight a fundamental difference: unlike free fermions, the effective dynamics of free bosons are governed by non-Hermitian pseudo-Hermitian dynamical matrices. Consequently, bosonic systems exhibit distinct dynamical stability phases (stable vs. unstable) separated by spectral singularities (Exceptional Points and Krein Collisions). The central problem is to determine if topological phase transitions in bosonic systems are inherently tied to critical correlations in the bosonic quasiparticle vacuum (QPV), or if these phenomena can be decoupled.

Methodology
The authors employ a combination of specific model construction and general mathematical arguments based on index theory:

Model Construction: They introduce a two-parameter family of translationally invariant QBHs derived from the bosonic Su-Schrieffer-Heeger (SSH) model. This model breaks particle-number conservation via a staggered pairing term while preserving a chiral pseudo-symmetry. The Hamiltonian is defined as an interpolation $H(2,0)(s) = sH(2) + (1-s)H(0)$ between two dimer QBHs, controlled by parameters $s$ (interpolation) and $\delta$ (pairing strength).
Spectral Analysis: The dynamics are analyzed via the $2n \times 2n$ $2 n \times 2 n$ pseudo-Hermitian dynamical matrix $G$ $G$ . The authors distinguish between:
- Dynamical Stability: Where $G$ is diagonalizable with a real spectrum.
- Dynamical Instability: Where $G$ has complex eigenvalues or is non-diagonalizable.
- Spectral Singularities: Exceptional Points (EPs) where eigenvalues and eigenvectors coalesce, and Krein Collisions (KCs) where eigenvalues of opposite Krein signature collide.
Topological Characterization:
- Berry Phase: They calculate a symplectic analogue of the Berry phase for the dynamically stable regime.
- Index Theory: To provide a robust classification that extends beyond dynamical stability, they utilize the index theorem for Toeplitz operators, linking the winding number of the bulk dynamical matrix to the number of boundary zero modes (ZMs).
Correlation Analysis: They compute the covariance matrix (CM) of the QPV to determine the range of correlations, investigating how these depend on the topological parameter $s$ versus the stability parameter $\delta$ .

Key Results

Decoupling of Topology and Criticality:
- The model exhibits a topological phase transition at $s = 1/2$ (mediated by Krein Collisions) and a dynamical instability transition at $\delta = 1$ (mediated by Exceptional Points). These two transitions are orthogonal in parameter space.
- Localization Length: The localization length of the boundary zero modes diverges as $s \to 1/2$ , signaling a topological transition. This divergence is entirely independent of $\delta$ .
- Correlation Length: Long-range correlations in the QPV arise only when the system approaches the dynamical instability line ( $\delta \to 1$ ) or when specific choices of the QPV are made. Crucially, the covariance matrix is independent of the topological parameter $s$ throughout the dynamically stable regime.
- Conclusion: A topological phase transition can occur (change in bulk invariant and boundary modes) without any associated criticality (divergence of correlation length) in the bulk.
Topological Classification Beyond Stability:
- The authors demonstrate that the chiral pseudo-symmetry of the model allows for a topological classification and a bulk-boundary correspondence even in the dynamically unstable regime ( $\delta > 1$ ).
- Using the index theorem, they show that the winding number of the bulk dynamical matrix predicts the existence of boundary zero modes regardless of whether the system is dynamically stable or unstable. The topological invariant remains well-defined even where the QPV is not unique or where the system is unstable.
Nature of the QPV:
- In the dynamically stable regime of this model, the Krein gap vanishes identically due to the degeneracy of bands with opposite Krein signatures. This results in a degenerate manifold of QPVs.
- Long-range correlations are "conditionally" available: they depend on the specific choice of QPV within this manifold, not on the topological transition itself.

Significance and Claims
The paper claims to provide a new understanding of criticality in bosonic systems described by quadratic Hamiltonians. Its primary significance lies in the decoupling of band topology from criticality:

Contrast with Fermions: Unlike fermionic systems where topological transitions are automatically associated with criticality (gap closing $\to$ diverging correlation length), bosonic systems can undergo topological phase transitions entirely within a single phase of dynamical stability, without generating long-range correlations.
Robustness of Topology: The topological features (bulk invariants and boundary zero modes) are shown to be insensitive to dynamical stability. The bulk-boundary correspondence holds even in dynamically unstable regimes where the standard notion of a ground state or unique QPV breaks down.
Theoretical Framework: The work suggests that bosonic topological physics, when extracted from basic index theory, is independent of dynamical stability and, consequently, independent of non-interacting criticality.

The authors conclude that while the link between bulk topology and boundary physics mirrors the fermionic case, the link between topological transitions and bulk criticality is broken in bosonic systems. This "decoupling" arises fundamentally from the non-Hermitian nature of bosonic dynamics and the specific structure of their pseudo-Hermitian dynamical matrices.

1. The Two Types of "Chaos"

2. The Great Decoupling

3. The "Ghost" of the Pattern

4. Why This Matters (According to the Paper)

Technical Summary: Decoupling Band Topology from Criticality in Bosonic Systems

More like this