Entanglement Asymmetry and Quantum Mpemba Effect for… — Plain-Language Explanation

✨

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 have a giant, perfectly organized library where every book is sorted by color, size, and author. This represents a universe with perfect symmetry. Now, imagine someone sneaks in and throws a few books onto the floor, mixing them up. The library is now "symmetry-broken."

Usually, if you leave a messy room alone, it stays messy or gets messier. But in the quantum world, there's a strange phenomenon called the Quantum Mpemba Effect. It's named after the idea that hot water can sometimes freeze faster than cold water. In our quantum library, this means: The messier the room starts out, the faster it cleans itself up.

This paper by Fujimura and Shimamori explores this "fast-cleaning" phenomenon in a very specific, complex type of quantum library called a Wess-Zumino-Witten (WZW) model. Here is the story of their discovery, broken down into simple concepts.

1. The Problem: Why Can't We Just Break Symmetry?

In our everyday world (3D space), you can easily break symmetry. Think of a magnet: all the tiny atomic arrows point one way, breaking the "no-direction" symmetry.

However, the authors are studying a world with only one dimension of space and one of time (a 1D line). There is a famous rule in physics called the Coleman-Mermin-Wagner theorem. It says: "In a 1D line, you cannot spontaneously break continuous symmetry." It's like trying to balance a pencil on its tip in a windy room; it will always fall back to the middle.

The Workaround: Since they can't break symmetry "naturally," they cheat. They create an excited state. Imagine taking a specific book (an operator) and throwing it onto the floor at a specific spot. This explicitly breaks the order. They then watch how the system tries to fix itself over time.

2. The Tool: Measuring the "Mess" (Entanglement Asymmetry)

How do you measure how messy the library is? The authors use a tool called Entanglement Asymmetry.

The Analogy: Imagine you are looking at just one shelf of the library (a subsystem).
Symmetry Preserved: If the books on that shelf are perfectly balanced and follow the rules, the "asymmetry" score is zero.
Symmetry Broken: If the books are jumbled, the score goes up.
The Goal: They want to see how fast this score drops back to zero as time passes.

3. The Discovery: The Standard Mpemba Effect

First, they confirmed the known phenomenon: The more you break the symmetry initially, the faster it restores.

Scenario A: You throw one book on the floor. It takes a while to clean up.
Scenario B: You throw ten books on the floor. Surprisingly, the system cleans up faster than in Scenario A.

It's counterintuitive. Usually, a bigger mess takes longer to clean. But in this quantum world, a "bigger" initial disturbance actually triggers a faster recovery mechanism.

4. The New Discovery: The "Rank" and "Level" Twist

This is where the paper gets really exciting. They didn't just look at how much they threw; they looked at what kind of library they were in. They changed two settings:

Rank ( $N$ ): Think of this as the size of the alphabet used to write the books. A higher rank means more types of books.
Level ( $k$ ): Think of this as the complexity of the rules governing how books can be stacked.

They found a brand new type of Mpemba effect that depends on these settings:

The "Rank" Effect (The Alphabet Size)

What happens: If you increase the Rank ( $N$ ), you are essentially making the initial mess worse (more symmetry is broken).
The Twist: Even though the mess is worse, the system cleans up faster.
Metaphor: Imagine a library with 100 different book genres vs. one with only 2. If you mess up the 100-genre library, it somehow snaps back to order more quickly than the small library, despite being more chaotic.

The "Level" Effect (The Rule Complexity)

What happens: If you increase the Level ( $k$ ), the initial mess becomes smaller (less symmetry is broken).
The Twist: Even though the mess is smaller, the system takes longer to clean up.
Metaphor: If you have very strict, complex rules for stacking books (high Level), even a tiny mistake takes a long time to fix because the system is "stiff."

Summary of the New Effect:

Higher Rank = Bigger Mess + Faster Cleanup.
Higher Level = Smaller Mess + Slower Cleanup.

5. The Adjoint Case: Not Everything is Universal

The authors also tested a different way of breaking symmetry (using "currents" instead of "primary operators").

Result: They saw the standard "messier = faster" effect.
But: They did not see the new "Rank/Level" effect.
Conclusion: This new type of fast-cleaning behavior isn't a universal law of the universe; it depends heavily on how you break the symmetry in the first place. It's like how some people clean up a mess quickly if they are angry (high energy), but others take their time even if the mess is small.

Why Does This Matter?

This paper is like finding a new rule in the game of quantum physics.

It challenges our intuition: We usually think "more work = more time." Here, "more work = less time."
It helps us understand quantum systems: By understanding how these systems "heal" themselves, we might get better at building quantum computers that can recover from errors faster.
It reveals hidden structures: The fact that changing the "Rank" of the theory changes the speed of recovery tells us that the fundamental structure of the universe (the number of dimensions or types of particles) is deeply connected to how time and order behave.

In short, the authors found that in the quantum world, sometimes the bigger the chaos, the quicker the order returns, but only if you are playing by the right rules and in the right kind of library.

1. Problem Statement

The paper addresses two interconnected challenges in non-equilibrium quantum many-body physics:

Quantifying Symmetry Breaking in Subsystems: Traditional order parameters (like magnetization) are ill-suited for characterizing symmetry breaking in non-equilibrium subsystems. The authors utilize Entanglement Asymmetry ( $\Delta S_A$ ), defined as the relative entropy between a reduced density matrix $\rho_A$ and its symmetrized version $\rho_{A,G}$ , to measure the degree of symmetry breaking at the subsystem level.
The Quantum Mpemba Effect (QME) for Non-Abelian Symmetries: The QME is the counterintuitive phenomenon where a system with more initial symmetry breaking restores symmetry faster than a system with less initial breaking. While observed for Abelian symmetries (e.g., $U(1)$ ) and in random circuits, the QME for Non-Abelian global symmetries (specifically $SU(N)$) in continuum field theories remains largely unexplored.
Theoretical Constraint: In $1+1$ dimensions, the Coleman-Mermin-Wagner theorem forbids spontaneous breaking of continuous global symmetries. To study symmetry restoration, the authors must consider excited initial states that explicitly break the symmetry, rather than relying on spontaneous symmetry breaking.

2. Methodology

The authors employ Conformal Field Theory (CFT) techniques, specifically the $\widehat{\mathfrak{su}}(N)_k$ Wess-Zumino-Witten (WZW) model, to analyze real-time dynamics.

Initial States: Two types of excited states are constructed by inserting local primary operators into the vacuum $|0\rangle$ $∣0 ⟩$ :
1. Fundamental Representation: $|\psi\rangle = \mathcal{O}_{fund} |0\rangle$ .
2. Adjoint Representation: $|\psi\rangle = J^a |0\rangle$ (constructed from WZW currents).
Entanglement Asymmetry Calculation:
- Direct calculation of $\Delta S_A$ is difficult due to the logarithm of the density matrix. The authors compute the $n$ -th Rényi Entanglement Asymmetry ( $\Delta S_A^{(n)}$ ) and take the limit $n \to 1$ .
- Using the replica trick, $\Delta S_A^{(n)}$ is expressed in terms of $2n$ -point correlation functions on an $n$ -sheeted Riemann surface.
- For $n=2$ (the focus of the paper), this reduces to a ratio of four-point functions.
Analytical Tools:
- Conformal Mapping: The $n$ -sheeted manifold is mapped to the complex plane using specific conformal transformations.
- Knizhnik-Zamolodchikov (KZ) Equations: The four-point functions for $\widehat{\mathfrak{su}}(N)_k$ are solved analytically using the KZ equations. The solutions involve hypergeometric functions and depend on the rank $N$ and the level $k$ .
- Symmetry Operators: The calculation involves integrating over the group manifold $SU(N)$ using Haar measures to project onto the invariant subspace, utilizing the topological nature of symmetry operators in CFT.

3. Key Contributions

First Systematic Study of Non-Abelian QME: The paper provides the first rigorous demonstration of the Quantum Mpemba effect for non-Abelian $SU(N)$ symmetry in a continuum field theory.
Discovery of a "New Type" of QME: Beyond the standard QME (where initial breaking strength is the control parameter), the authors identify a novel dependence on the theory's parameters:
- Rank Dependence ( $N$ ): Increasing the rank $N$ (which increases the initial symmetry breaking) leads to faster symmetry restoration.
- Level Dependence ( $k$ ): Increasing the level $k$ (which decreases initial symmetry breaking) leads to slower symmetry restoration.
Representation Dependence: The authors demonstrate that this "new type" of QME is not universal; it appears in the fundamental representation but is absent in the adjoint representation.

4. Detailed Results

A. Fundamental Representation Case

Standard QME: By varying the insertion distance $\tau_0$ (controlling initial breaking), the authors confirmed that states with larger initial $\Delta S_A^{(2)}$ decay faster, exhibiting curve crossings in time evolution.
New Type QME (Varying $N$ ):
- Initial State: For small $\tau_0$ , the initial asymmetry scales as $\log[N(N+1)/2]$ (for specific insertion points), increasing with $N$ .
- Long-Time Dynamics: The decay rate of asymmetry is governed by the conformal dimension and level.
- Result: Curves for different $N$ (where $N < k$ ) cross. Higher $N$ starts with higher asymmetry but decays to zero faster. This implies that a "more broken" state (higher $N$ ) restores symmetry more rapidly.
New Type QME (Varying $k$ ):
- Initial State: Increasing $k$ reduces the initial asymmetry.
- Long-Time Dynamics: Increasing $k$ slows down the decay rate.
- Result: Higher $k$ starts with lower asymmetry but takes longer to restore symmetry.
Special Case ( $N=k$ ): When the rank equals the level, the behavior changes. For certain insertion points, double crossings are observed, indicating a more complex restoration structure.

B. Adjoint Representation Case

The authors analyzed the initial state built from WZW currents ( $J^a$ ).
Standard QME: The standard QME (varying $\tau_0$ ) is still observed; curves cross, showing faster restoration for more broken states.
Absence of New Type QME: When varying $N$ , the initial asymmetry increases with $N$ , but the long-time decay rate also increases (faster decay) in a way that prevents curve crossings. The inequality $\Delta S_A(t=0, N) < \Delta S_A(t=0, N')$ and $\Delta S_A(t=\infty, N) < \Delta S_A(t=\infty, N')$ holds simultaneously, meaning the ordering of curves never flips. Thus, the "new type" QME is specific to the fundamental representation.

C. Asymptotic Behavior

Long-Time Limit ( $t \to \infty$ ): $\Delta S_A^{(2)}$ decays to zero as a power law determined by $N$ and $k$ . The leading term scales as $\epsilon^{2N/(N+k)}$ (for $N<k$ ) or $\epsilon$ (for $k<N$ ), where $\epsilon \sim (\tau_0/t^2)^2$ .
Large Interval Limit ( $\tau_0 \to 0$ ): The dynamics are well-described by a quasi-particle picture. Symmetry breaking propagates via left- and right-moving quasi-particles. The entanglement asymmetry is non-zero only when these particles are within the subsystem $A$ . The magnitude of the plateau is determined by the dimension of the representation (e.g., $\log N$ for fundamental, $\log(N^2-1)$ for adjoint).

5. Significance and Implications

Universality of QME: The work challenges the notion that the QME is a universal feature of all symmetry-breaking scenarios. The dependence on the representation (fundamental vs. adjoint) suggests that the microscopic structure of the excitation plays a crucial role in the restoration dynamics.
Parameter-Driven QME: The discovery that varying the theory parameters ( $N$ and $k$ ) rather than just the initial state can induce a QME opens a new direction in studying non-equilibrium dynamics. It links the "speed" of symmetry restoration directly to the algebraic structure of the underlying CFT.
Theoretical Framework: The paper establishes a robust analytical framework using WZW models and KZ equations to study non-Abelian symmetry restoration, which can be extended to other affine Lie algebras (e.g., $SO(N)$, $Sp(N)$).
Future Directions: The authors suggest extending these results to finite temperature, calculating the $n \to 1$ limit (relative entropy) directly, and interpreting the results via the holographic dual (3D Chern-Simons theory).

In conclusion, this paper provides a rigorous proof of the Quantum Mpemba effect for non-Abelian symmetries and uncovers a novel, representation-dependent mechanism where the rank of the symmetry group dictates the speed of symmetry restoration.

Entanglement Asymmetry and Quantum Mpemba Effect for Non-Abelian Global Symmetry