⚛️ quantum physics

Microscopic study of topological phase transitions: Percolation point of view

This paper investigates the microscopic mechanisms of decoherence-induced topological phase transitions in color and toric codes by interpreting topological entanglement negativity and disorder parameters through a percolation framework, revealing that while quasi-local entanglement negativity effectively visualizes decohered regions, the distinct responses of these models to biased decoherence patterns indicate that a universal microscopic description of such transitions remains elusive.

Original authors: Keisuke Kataoka, Ikuo Ichinose

Published 2026-07-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Keisuke Kataoka, Ikuo Ichinose

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 the universe as a giant, invisible game of connect-the-dots, but instead of drawing lines with a pencil, the universe uses a mysterious force called "quantum entanglement." In this game, particles are linked in such a way that they act like a single team, no matter how far apart they are. This is the world of topological order, a special state of matter that physicists love because it's incredibly tough to break. Think of it like a knot in a piece of string: you can shake the string, pull it, or twist it, but the knot stays tied unless you cut the string. This "knotting" is what makes quantum computers potentially powerful and error-resistant.

However, the real world is messy. Things like heat, noise, or just interacting with the environment cause decoherence, which is like someone trying to untie that knot by shaking the string too hard. When this happens, the special "knot" can unravel, and the material changes from a super-robust quantum state to a boring, ordinary one. The big question scientists are asking is: Exactly how does this unraveling happen? Does the knot loosen slowly and evenly, or does it suddenly snap apart? And do the different ways we measure the "knot" (like checking how tangled the string is versus checking if the string is still one piece) tell the same story?

This paper dives into that messy middle ground. The authors, Keisuke Kataoka and Ikuo Ichinose, act like microscopic detectives, zooming in to watch how a quantum state breaks down when it gets "decohered." They use two famous quantum models, the Color Code and the Toric Code, as their test subjects. Instead of just looking at the whole system, they invent a new tool called Quasi-Local Topological Entanglement Negativity (QLTEN). You can think of QLTEN as a high-tech thermal camera that lets them take snapshots of the quantum state, showing exactly where the "knots" are still holding strong and where they have already turned into a tangled mess.

They also introduce a twist to their experiment called Explosive Percolation. Imagine you are building a bridge by dropping planks one by one. In a normal scenario, you drop them randomly, and the bridge grows slowly. In "explosive" percolation, you have a rule that prevents you from dropping a plank if it would connect two huge groups of planks too early. This forces the bridge to stay broken into many small islands for a long time, until—snap!—one giant bridge suddenly appears. The authors use this to see if the quantum state behaves differently when the "noise" (decoherence) is forced to grow in this specific, biased way.

Here is what they found:

First, they confirmed that as the noise increases, the Color Code (their first test subject) does indeed transform into a Toric Code state before eventually losing its quantum magic entirely. They watched this happen by tracking two things: the Topological Entanglement Negativity (TEN), which measures how "knotted" the system is, and the disorder parameter, which checks if the system's symmetry is broken.

However, the story gets interesting when they look at the details. They discovered that while the global numbers (the average TEN and the symmetry) seemed to change at a specific point, the local snapshots told a different story. Using their new QLTEN camera, they saw that the transition wasn't a smooth, uniform slide. Instead, it looked like a landscape where islands of the new "Toric Code" state started popping up inside the old "Color Code" state, like droplets of water forming in a dry desert.

When they applied the Explosive Percolation rule (the "biased noise"), they found a surprising difference between their two test subjects. In the Color Code, the system was surprisingly stubborn. Even when the noise was forced to grow in this explosive, clustered way, the quantum state didn't seem to care much. The "knots" held on, and the transition point didn't shift much. It was as if the Color Code was wearing a super-strong suit of armor that ignored the specific pattern of the attack.

But the Toric Code was a different story. When they applied the same explosive noise, the Toric Code reacted very differently. The "Higgs region" (the part where the quantum order is lost) grew in a way that was tightly linked to the explosive clusters. The logical qubits (the information stored in the knot) survived much longer than expected, hiding inside the growing clusters until the very last moment.

The paper suggests that this difference is crucial. It implies that we cannot assume all quantum systems will break down in the same way, even if they seem similar on the surface. The Color Code and the Toric Code, while related, have different "personalities" when it comes to handling noise.

One of the most exciting findings is how the QLTEN snapshots acted as an early warning system. In the Toric Code, the QLTEN showed that the system was almost entirely in the "broken" state (the Higgs region) before the logical qubits actually disappeared. This means that by looking at these local snapshots, we might be able to predict when a quantum computer is about to lose its data, giving us a chance to fix it before it's too late.

In short, the authors didn't just watch the quantum knot untie; they mapped the exact shape of the unraveling. They found that the process is messy, local, and surprisingly different depending on which quantum system you are looking at. Their work suggests that understanding these microscopic details is the key to building better, more reliable quantum computers that can survive the noisy real world.

Technical Summary: Microscopic Study of Topological Phase Transitions: Percolation Point of View

Problem Statement
The paper investigates the microscopic mechanisms governing topological phase transitions, specifically how a topologically ordered state evolves into another topological state or a non-topological state under quantum decoherence. While topological order is characterized by two distinct but correlated aspects—quantum entanglement and 1-form symmetry—current understanding lacks a complete picture of how these aspects intertwine during the critical regime of a transition. Key open questions include whether indicators of entanglement (e.g., Topological Entanglement Entropy, TEE) and 1-form symmetry exhibit singular behavior at the same critical point with identical critical exponents, and how these phenomena can be understood from a microscopic, geometric perspective.

Methodology
The authors employ two primary case studies:

Color Code to Toric Code (TC) Transition: A honeycomb lattice color code is subjected to "red XX-decoherence," modeled as a stochastic process where $XX$ operators are applied to red links with probability $p$ . This induces the proliferation of $rX$-anyons, effectively transforming the system into a TC on an emergent triangular lattice.
Toric Code to Higgs Phase Transition: A TC on a triangular lattice is subjected to $Z$ -decoherence (external field type), driving a transition to a Higgs phase.

Key Observables and Tools:

Stabilizer Formalism: The state evolution is tracked via changes in the stabilizer group. Decoherence is treated as the merging of stabilizers and the discarding of specific operators, corresponding to anyon proliferation.
Topological Entanglement Negativity (TEN): Used as a measure of quantum entanglement for mixed states, calculated via the rank of a matrix derived from truncated stabilizer generators.
Disorder Parameter of 1-form Symmetry: Defined as the expectation value of a string operator, used to detect the restoration of weak 1-form symmetry.
Quasi-Local TEN (QLTEN): A novel observable introduced to map the local topological properties of the system. By calculating TEN on small, overlapping 7-plaquette hexagonal subsystems, the authors generate "snapshots" of the system, visualizing the geometric distribution of topological phases (e.g., TC regions vs. trivial regions).
Explosive Percolation (EP): To probe the robustness of these transitions, the authors introduce biased decoherence using the "Achlioptas process" (specifically the product rule). This suppresses the formation of large decoherence clusters until a critical threshold, allowing for a comparison between standard random percolation and EP dynamics.
Simplicial Homology: The analysis interprets the disorder parameter and TEN through the lens of simplicial homology, identifying them as 0-th and 1-st homological objects, respectively.

Key Results

Divergence of Critical Behaviors: In the Color Code transition, the TEN (entanglement) and the disorder parameter (1-form symmetry) exhibit distinct behaviors. The variance of TEN peaks at $p \approx 0.22$ , while the disorder parameter indicates a continuous phase transition with a critical point $p_c \approx 0.31$ and critical exponent $\nu \approx 1.2$ . This discrepancy suggests that entanglement and symmetry restoration do not necessarily occur simultaneously or with the same criticality.
Geometric Interpretation via QLTEN: QLTEN snapshots reveal that the transition is not uniform. TC regions emerge as "droplets" within the color code phase. The study finds that a reduction in global TEN from 2 to 1 occurs when at least one of the three subsystems in the calculation forms a (quasi-)closed loop of decohered links. This links the global topological invariant to local percolation properties.
Homological Distinction: The paper posits that the observed discrepancy arises because the disorder parameter measures 0-th homology (connected clusters), while TEN measures 1-st homology (closed loops). These topological features evolve differently under decoherence.
System-Dependent Response to Biased Decoherence:
- Color Code: The system exhibits remarkable robustness. The global TEN, disorder parameter, and QLTEN cluster sizes are largely insensitive to the type of percolation (standard vs. EP). The critical thresholds remain similar regardless of the bias in cluster growth.
- Toric Code: In contrast, the TC is highly sensitive to EP. As the EP parameter $n$ increases, the emergence of the Higgs phase is significantly delayed. The TEN and logical qubit survival show strong dependence on the percolation pattern.
Logical Qubit Survival: In the TC study, the authors track the survival of non-contractible logical operators ('tHooft loops). They find that the logical qubits persist even when the system is largely in the Higgs phase ( $\gamma_N = 0$ ), provided the largest QLTEN cluster has not yet spanned the entire system. The disappearance of logical qubits coincides with the percolation of the Higgs region across the system.

Significance and Claims
The paper claims that the introduction of QLTEN provides a crucial microscopic and geometric view of topological phase transitions, bridging the gap between global topological invariants and local percolation phenomena. By framing the transition through simplicial homology, the authors offer a theoretical explanation for the distinct behaviors of entanglement and symmetry indicators.

A primary finding is the lack of universality in topological phase transitions: even though the Color Code and TC are closely related (the former being a product of two TCs), they react fundamentally differently to biased decoherence patterns. This suggests that obtaining a universal understanding of topological phase transitions is challenging and likely model-dependent.

Furthermore, the work highlights the potential of QLTEN for quantum error mitigation. Since QLTEN configurations (derived from stabilizer syndromes) can predict the percolation of the trivial phase before logical qubits are destroyed, they may serve as an early warning system for the failure of topological quantum codes. The paper concludes that while QLTEN and other observables correlate in standard scenarios, their behavior under biased conditions reveals deep, system-specific nuances in how topological order degrades.

More like this