🔬 mesoscale physics

Scaling behavior of eigenspectrum for entanglement from correlation matrices

This paper elucidates the logarithmic scaling of entanglement entropy in one-dimensional critical systems by analytically demonstrating that the eigenvalues of correlation matrices scale as $1/L_A$ in the bulk, a result derived via a circulant matrix approximation that bridges the gap between eigenvalue distributions and universal entanglement behavior.

Original authors: Lih-King Lim

Published 2026-09-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Lih-King Lim

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

In the quantum world, particles do not exist in isolation; they are bound together by a strange and powerful connection known as entanglement. When two parts of a system are entangled, the state of one cannot be described without the other, even if they are separated by vast distances. This phenomenon is not just a theoretical curiosity; it is the fundamental resource behind quantum computing and offers deep insights into how matter organizes itself at the smallest scales. Physicists have long been fascinated by how this connection changes as the size of a system grows. In many everyday materials, the amount of entanglement between a piece of the system and the rest grows in proportion to the surface area of that piece. However, in a special class of materials known as critical systems—those sitting right at the edge of a phase transition, like water freezing into ice but not quite there—the rules change. Here, the entanglement grows much faster, following a pattern that depends on the logarithm of the system's size. This behavior is a hallmark of criticality, but understanding exactly why it happens has traditionally required extremely complex mathematics, often involving advanced theories that are difficult to visualize or intuitively grasp.

A researcher at Zhejiang University has now peeled back the layers of this complexity by looking at the problem from a different angle. Instead of relying on the heavy machinery of conformal field theory or intricate asymptotic formulas, they focused on the individual building blocks that make up the entanglement: the numbers that describe how likely different quantum states are to occur. These numbers, which come from a mathematical object called a correlation matrix, act like a map of the system's internal connections. The researcher discovered that for both free-moving particles on a lattice and a specific type of magnetic chain known as the transverse field Ising model, these numbers follow a remarkably simple rule when the system is at a critical point. As the size of the subsystem increases, the values of these numbers in the middle of the spectrum shrink in direct proportion to the size of the system. Specifically, they become smaller by a factor related to the inverse of the subsystem size.

To see this clearly, the scientist replaced the original, complicated mathematical structure with a simpler, circular version that behaves almost identically when the system is large. This trick allowed them to calculate the behavior of each individual number without getting lost in the algebra. They found that in the critical state, most of these numbers cluster near the extremes of zero and one, representing states that are either completely empty or completely full. However, right in the middle, where the transition happens, the numbers scale down steadily. When you add up the contribution of all these numbers to calculate the total entanglement, this steady shrinking leads directly to the famous logarithmic growth. It is as if the total amount of connection is built from a vast number of tiny, diminishing pieces, and the sheer number of these pieces growing with the system size creates the logarithmic curve. This finding provides a clear, mechanical explanation for a result that was previously understood only through abstract theory.

The study did not stop at the critical point. The researcher also explored what happens when the system is pushed away from this delicate balance, into a non-critical state where entanglement is usually expected to be minimal and constant. They started from a limit where the magnetic field is so strong that all spins align perfectly, creating a state with no entanglement at all. As they gradually weakened this field, they observed a smooth transition. Even in this non-critical regime, the numbers describing the system still showed a scaling behavior similar to the critical case, shrinking as the system grew larger. However, the rate of this shrinkage was incredibly slow, governed by a tiny factor that depends on how far the system is from the critical point. This subtle effect means that the entanglement still grows logarithmically with size, but so gently that it appears almost constant to the naked eye. This explains why non-critical systems seem to have a fixed amount of entanglement, while critical systems show a dramatic increase.

By focusing on the individual values that make up the correlation matrix, the researcher has offered a new way to visualize the architecture of quantum entanglement. They showed that the robust logarithmic scaling seen in critical systems is not a mysterious property of the whole, but a direct consequence of how the individual components of the system scale down as the system expands. This approach bypasses the need for the most sophisticated mathematical tools, replacing them with a more intuitive picture of how these quantum connections are distributed. The work confirms that the scaling behavior is a fundamental feature of the eigenvalues themselves, present even when the system is perturbed away from the critical point, though the effect becomes vanishingly small. This insight bridges the gap between the abstract mathematics of quantum field theory and the concrete behavior of individual quantum states, providing a clearer path to understanding how complexity emerges from simplicity in the quantum realm.

Technical Summary: Scaling Behavior of Eigenspectrum for Entanglement from Correlation Matrices

Problem Statement
The paper addresses the scaling behavior of entanglement entropy in one-dimensional critical quantum systems, specifically focusing on the ground states of free lattice fermions and the transverse field Ising chain. While it is a well-established result that the von Neumann entanglement entropy ( $S_{vN}$ ) of these systems scales logarithmically with the subsystem size ( $L_A$ ) as $S_{vN} \sim (c/3) \log_2 L_A$ (where $c$ is the conformal charge), the derivation of this result typically relies on sophisticated mathematical physics techniques, such as the Fisher-Hartwig formula for Toeplitz determinants, or conformal field theory (CFT). The author aims to provide a more intuitive, underlying perspective on this scaling by analyzing the distribution of individual eigenvalues of the correlation matrix, rather than relying solely on the total behavior of the spectrum or determinant.

Methodology
The author employs a circulant matrix approximation to analyze the correlation matrices of the target systems.

Correlation Matrices: For solvable models (non-interacting fermions and the transverse field Ising chain), the reduced density matrix is determined by the correlation matrix $C_A$ . Due to translation invariance, $C_A$ takes the form of a Toeplitz (or block-Toeplitz for the Ising chain) matrix.
Circulant Approximation: Instead of directly solving the Toeplitz problem, the author constructs an asymptotically equivalent circulant matrix. In the limit of large matrix dimension ( $L_A \to \infty$ ), the total behavior of the eigenvalues of the circulant matrix is equivalent to that of the original Toeplitz matrix.
Analytical Derivation: The circulant structure allows for a straightforward analytical expression of eigenvalues via discrete Fourier transforms. The author converts these sums into integrals in the large- $L_A$ limit to derive asymptotic expressions for the eigenvalues $\lambda_m$ .
Scaling Analysis: The author analyzes the scaling of individual eigenvalues in the "bulk" of the spectrum (away from the edges) and subsequently sums the contributions to the entropy function $S_{vN} = -\sum [\lambda_m \ln \lambda_m + (1-\lambda_m) \ln (1-\lambda_m)]$ .
Non-Critical Extension: The methodology is extended to the non-critical regime of the Ising chain by perturbing from the entanglement-free limit ( $h = \infty$ ) using a phase profile parameterization involving a sawtooth function and Bessel functions.

Key Contributions and Results

Eigenvalue Scaling in Critical Systems:
- For both free lattice fermions and the critical transverse field Ising chain ( $h=1$ ), the author analytically demonstrates that the eigenvalues in the bulk of the eigenspectrum scale as $1/L_A$ .
- Specifically, eigenvalues cluster near 0 or 1, with the transition region scaling as $\lambda_m \approx 1 - \delta$ or $\delta$ , where $\delta \propto 1/L_A$ .
- When substituted into the entropy function, terms of the form $-\delta \ln \delta$ yield a contribution proportional to $\ln L_A$ . Summing over the $L_A$ terms in the bulk reproduces the logarithmic scaling of the entanglement entropy ( $S_{vN} \sim \ln L_A$ ).
- This result explains the logarithmic scaling from the perspective of individual eigenvalue behavior, a detail not explicitly evident in the Fisher-Hartwig formula which focuses on the total determinant.
Non-Critical Scaling and Crossover:
- The author investigates the Ising chain away from the critical point, starting from the entanglement-free limit ( $h \to \infty$ ).
- They find that the $1/L_A$ scaling of the eigenvalues persists even in the non-critical regime.
- However, the magnitude of the eigenvalues is suppressed by a small parameter $\beta^2$ (related to the deviation from the critical point).
- This results in a "smooth crossover" to non-critical scaling characterized by a very weak logarithmic dependence: $S_{vN} \sim \beta^2 \log_2 L_A$ . As $\beta \to 0$ (approaching the product state limit), the entanglement vanishes, appearing effectively constant for finite subsystem sizes, consistent with expectations for non-critical systems.
Numerical Verification:
- The analytical results derived from the circulant approximation are compared with exact numerical calculations of the Toeplitz matrix eigenvalues.
- While the circulant approximation reproduces the correct logarithmic dependence, the author notes that the slope (the rate of divergence) differs from the exact microscopic or CFT universal coefficients. This discrepancy is attributed to the fact that the entropy function involves a non-analytic $x \log x$ term at the origin, which may require stricter conditions than simple asymptotic equivalence of matrices.

Significance and Claims
The paper claims to offer a new, intuitive perspective on the robust logarithmic scaling of entanglement entropy at criticality. By shifting the focus from the global properties of Toeplitz determinants to the scaling of individual eigenvalues in an asymptotically equivalent circulant matrix, the author provides a simplified analytical framework that avoids the technical complexity of the Fisher-Hartwig formula.

The author modestly states that while their method captures the functional form of the scaling ( $\log L_A$ ), it does not reproduce the exact universal coefficients (e.g., $c/3$ ) found in CFT or exact microscopic studies. They attribute this to the limitations of the asymptotic equivalence regarding the non-analytic nature of the entropy function near zero. Nevertheless, the work successfully demonstrates that the $1/L_A$ scaling of bulk eigenvalues is the fundamental mechanism driving the logarithmic entanglement growth in these systems. The method is also presented as a potential tool for studying other Toeplitz-related problems, such as mid-spectrum eigenstate entanglement and random Toeplitz matrices.

More like this