🔢 mathematics

Entanglement entropy of ground states of the Landau Hamiltonian on the half-plane

This paper demonstrates that despite the Landau Hamiltonian on the half-plane possessing a purely absolutely continuous spectrum (unlike the full plane's pure point spectrum), its ground states still obey a strict area law for entanglement entropy, prompting an investigation into the specific spectral conditions required to produce a logarithmically enhanced area law.

Original authors: Paul Pfeiffer, Wolfgang Spitzer

Published 2026-07-13

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Paul Pfeiffer, Wolfgang Spitzer

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 have a giant, invisible dance floor where electrons are the dancers. These dancers are "fermions," which means they are incredibly shy and refuse to stand on the same spot as anyone else. When the music stops (the system reaches its lowest energy state, or "ground state"), the dancers settle into a specific pattern.

Physicists love to measure how "entangled" these dancers are. Think of entanglement as a measure of how much the dancers on one side of the room are secretly communicating with the dancers on the other side. Usually, if you look at a chunk of this dance floor, the amount of secret communication (called Entanglement Entropy) depends on the size of the chunk.

For a long time, scientists noticed a pattern: if the dancers were in a room with no walls (the "full plane"), the communication usually scaled with the surface area of the chunk's boundary. This is called the "Area Law." However, there was a twist. Sometimes, if the dancers were moving in a very specific, fluid way (called "absolutely continuous spectrum"), the communication didn't just grow with the area; it grew with the area plus a little extra logarithmic boost. It was like the dancers were whispering a bit louder than expected.

The Big Question
The authors of this paper, Paul Pfeiffer and Wolfgang Spitzer, decided to test this theory in a very specific, tricky room: a half-plane. Imagine a dance floor that stops abruptly at a wall (the boundary at $x_1 = 0$ ). They put a strong magnetic field on this floor, which forces the electrons to dance in tight circles.

Here is the surprise: In this half-plane room, the electrons have a "purely absolutely continuous spectrum." Based on previous rules, scientists expected this to mean the entanglement would get that extra logarithmic boost. They thought the boundary might change the music so much that the "Area Law" would break and become an "Enhanced Area Law."

The Discovery
The paper proves that this expectation is wrong.

Even though the electrons in this half-plane room are dancing in that "fluid," continuous way that usually causes the extra boost, the entanglement entropy still follows the strict Area Law. It scales perfectly with the length of the boundary, with no extra logarithmic factor.

To put it simply: The wall (the boundary condition) changes the type of music the electrons dance to (changing the spectrum from discrete points to a continuous flow), but it does not change the rule of how much they talk to each other. The "strict area law" holds firm.

How They Figured It Out
The authors didn't just guess; they did the math. They compared the half-plane room to a full, infinite room.

They used a mathematical tool called a "Fourier transform" to break the problem down into smaller, manageable pieces (like analyzing the music note by note).
They calculated the difference between the electron patterns in the half-plane and the full plane.
They found that the difference between the two patterns decays very quickly (exponentially) as you move away from the wall.

Because this difference dies out so fast, the "noise" the wall introduces isn't strong enough to create that extra logarithmic boost. The math shows that for any region $\Lambda$ that is far enough away from the wall, the entanglement entropy is exactly the same as if the wall didn't exist at all.

What They Didn't Solve
The paper is very careful to say what they didn't find. They proved that a purely continuous spectrum does not guarantee a logarithmic boost. However, they leave an open question: What exact conditions are needed to get that boost?

They point out that in a 3D version of this problem (a long tube with a magnetic field), the electrons do get that logarithmic boost. So, the difference between the 2D half-plane (no boost) and the 3D tube (boost) is a mystery they haven't fully cracked yet. They know the electrons in the 3D tube have a "mixed" decay (fast in some directions, slow in others), and that slow decay seems to be the culprit for the boost. But in the 2D half-plane, even though the decay is slow in one direction, it's not enough to break the strict area law.

The Bottom Line
This paper is a solid proof (not just a simulation or a guess) that having a "fluid" spectrum isn't enough to break the Area Law. The boundary conditions in this specific 2D magnetic setup are too polite to let the entanglement get that extra boost. The authors have successfully shown that the "Strict Area Law" is more robust than we thought, surviving even when the spectrum changes drastically. But they also admit that the full recipe for when the "Logarithmic Boost" appears is still a mystery waiting to be solved.

Technical Summary: Entanglement Entropy of Ground States of the Landau Hamiltonian on the Half-Plane

Problem Statement
The paper investigates the scaling behavior of the entanglement entropy (EE) for ground states of non-interacting fermionic systems defined on a spatial domain with a boundary. Specifically, it focuses on the Landau Hamiltonian defined on the half-plane $\Omega = \mathbb{R}_+ \times \mathbb{R}$ with Dirichlet boundary conditions. The central question is how the presence of a boundary and the resulting spectral properties affect the leading-order asymptotic scaling of the EE as the system size $L$ tends to infinity.

Previous literature has established a dichotomy in EE scaling:

Strict Area Law: Scaling as $L^{d-1}$ (where $d$ is the spatial dimension). This is typically associated with systems having a pure point spectrum (e.g., the Landau Hamiltonian on the full plane $\mathbb{R}^2$ ) or exponentially decaying integral kernels of the spectral projection.
Logarithmically Enhanced Area Law: Scaling as $L^{d-1} \ln(L)$ . This is observed in systems with purely absolutely continuous spectrum and slow decay of the spectral projection kernel (e.g., the Laplace operator on $\mathbb{R}^d$ ).

The Landau Hamiltonian on the full plane ( $H$ ) has a pure point spectrum (Landau levels) and exhibits a strict area law. In contrast, the Landau Hamiltonian on the half-plane ( $H_+$ ) possesses a purely absolutely continuous spectrum. The authors aim to determine whether this change in spectral nature (from pure point to absolutely continuous) induces a transition to a logarithmically enhanced area law, or if the strict area law persists despite the continuous spectrum.

Methodology
The authors employ a rigorous operator-theoretic approach involving the following steps:

Direct Integral Decomposition: Using a partial Fourier transform with respect to the unbounded coordinate $x_2$ , the Hamiltonian $H_+$ is decomposed into a direct integral of one-dimensional shifted harmonic oscillators, $H_+(k) = -\frac{d^2}{dx^2} + (x-k)^2$ , defined on $L^2(\mathbb{R}_+)$ with Dirichlet boundary conditions at $x=0$ . The full-plane Hamiltonian $H$ is similarly decomposed into oscillators $H(k)$ on $L^2(\mathbb{R})$ .
Spectral Analysis: The ground state projections $P_\mu = \mathbb{1}(H \le \mu)$ and $P_\mu^+ = \mathbb{1}(H_+ \le \mu)$ are analyzed. For the half-plane, the spectrum of $H_+(k)$ depends on the parameter $k$ . The authors identify a critical value $k_F$ such that for $k < k_F$ , the ground state energy exceeds the Fermi energy $\mu$ , while for $k \ge k_F$ , a single eigenstate exists below $\mu$ .
Comparison of Projections: The core of the proof involves estimating the difference between the localized Fermi projections on the half-plane and the full plane, specifically the operator $D = P_\mu^+ - P_\mu(\mathbb{R}_+ \times \mathbb{R})$ . The authors derive an explicit integral representation for the Hilbert-Schmidt norm of this difference restricted to unit squares.
Estimates on Parabolic Cylinder Functions: The eigenfunctions of the shifted oscillators are expressed in terms of parabolic cylinder functions $D_\nu$ . The authors derive precise pointwise and $L^2$ -norm estimates for the difference between the eigenfunctions on the half-line ( $\psi_k$ ) and the full line ( $\phi_k$ ). These estimates show that the difference decays exponentially with respect to the shift parameter $k$ .
Schatten Norm Bounds: By summing the local estimates over the domain and utilizing the $p$ -triangle inequality for Schatten $p$ -quasi-norms (specifically for $p \in (0, 1]$ ), the authors bound the global norm of the difference operator.
Trace Inequalities: Finally, they apply a Sobolev-type inequality for functions of operators to bound the difference in entanglement entropies $|S_\alpha^+(L\Lambda) - S_\alpha(L\Lambda)|$ by the Schatten norm of the difference of the projections.

Key Results

Strict Area Law for Half-Plane: The main theorem (Theorem 2.1) proves that for any bounded $C^3$ -smooth region $\Lambda \subset \mathbb{R}_+ \times \mathbb{R}$ and Fermi energy $\mu \in (1, 3)$ , the $\alpha$ -Rényi entanglement entropy $S_\alpha^+(L\Lambda)$ satisfies a strict area law bound:
$S_\alpha^+(L\Lambda) \le C L |\partial \Lambda|$
where $C$ is independent of $L$ and $\Lambda$ .
Asymptotic Equivalence: If the domain $\Lambda$ is strictly separated from the boundary $\{0\} \times \mathbb{R}$ , the leading order term of the entropy on the half-plane is identical to that on the full plane:
$S_\alpha^+(L\Lambda) = L |\partial \Lambda| M_0(h_\alpha) + o(L)$
This implies that the Dirichlet boundary condition does not alter the leading-order scaling of the EE, even though it fundamentally changes the spectrum from pure point to absolutely continuous.
Decay Properties: The integral kernel of the Fermi projection $P_\mu^+$ exhibits exponential decay in the direction parallel to the boundary ( $x_1$ ) but only polynomial decay ( $O(1/|t|)$ ) in the direction perpendicular to the boundary ( $x_2$ ).

Significance and Claims
The paper claims to provide a counter-example to the heuristic that a purely absolutely continuous spectrum necessarily leads to a logarithmically enhanced area law. While the Laplace operator (purely absolutely continuous) exhibits the enhanced law, the Landau Hamiltonian on the half-plane (also purely absolutely continuous) exhibits a strict area law.

The authors conclude that the presence of absolutely continuous spectrum is not a sufficient condition for a logarithmically enhanced area law. They highlight that the decay properties of the integral kernel are crucial: while the half-plane projection decays exponentially in one direction, its polynomial decay in the other is insufficient to generate the logarithmic enhancement observed in the Laplacian case (where decay is polynomial in all directions).

The paper ends by posing an open question: What specific, finer conditions on the spectral measure or the decay of the integral kernel of a Schrödinger operator are necessary to guarantee a logarithmically enhanced area law? The authors do not provide a definitive answer to this question but suggest that the interplay between spectral type and spatial decay is more subtle than previously assumed.

More like this