Wave operators for Jacobi matrices

This paper establishes the existence and completeness of wave operators for Jacobi matrices whose spectral measures satisfy the Szegő condition, provided a mild additional assumption holds for the associated Verblunsky coefficients.

The Gist

The Big Picture: A Symphony of Chaos and Order

Imagine a giant, infinite piano stretching out forever to the right. Each key on this piano represents a number, and the way the keys are tuned determines how sound travels through the instrument. In mathematics, this instrument is called a Jacobi Matrix.

The authors of this paper, Sergey Denisov and Giorgio Young, are asking a fundamental question: If we pluck a string on this piano (start a vibration), how does the sound travel over time?

Specifically, they want to know: Does the sound eventually spread out smoothly across the whole piano like it would on a perfectly tuned, "free" instrument? Or does it get stuck, trapped in a corner, or behave in a chaotic, unpredictable way?

The Characters in Our Story

1. The Free Piano ($J_0$): Imagine a perfect piano where every key is exactly the same distance apart and tuned perfectly. If you play a note here, the sound wave travels perfectly smoothly and predictably. This is the "free evolution."
2. The Real Piano ($J$): Now, imagine our piano has some imperfections. Maybe some keys are slightly heavier, or the spacing is a tiny bit off. These imperfections are the "perturbations."
3. The Wave Operators ($\Omega$): These are the mathematical tools the authors use to compare the Real Piano to the Free Piano. They ask: "Can we translate the messy, chaotic sound of the Real Piano into the clean, smooth sound of the Free Piano?"
   • If the answer is Yes, the wave operators "exist."
   • If the answer is Yes, and we can translate every possible sound, the wave operators are "complete."

The Main Challenge: The "Szegő Condition"

The authors are studying a specific type of imperfect piano where the imperfections aren't random chaos. They follow a rule called the Szegő condition.

Think of this condition as a "quality control" rule. It says: "The imperfections on the piano are small enough that, on average, the piano still behaves like a good instrument." It's like saying, "Even though a few keys are slightly out of tune, the overall melody is still recognizable."

However, just being "mostly good" isn't always enough to guarantee the sound travels perfectly. Sometimes, even small imperfections can accumulate and trap the sound.

The New Discovery: The "Logarithmic Speed Limit"

The big breakthrough in this paper is finding a specific rule that guarantees the sound will travel smoothly.

The authors prove that if the imperfections on the piano get smaller fast enough, the sound will behave perfectly. But they found a very precise way to measure "fast enough."

They introduced a condition involving a logarithm (a slow-growing number).

• Imagine you are walking down a long hallway (the piano keys).
• You are checking the "roughness" of the floor at every step.
• The authors say: "As long as the average roughness of the floor, multiplied by the logarithm of how far you've walked, goes to zero, the sound will travel freely."

The Analogy:
Think of the imperfections as pebbles on a road.

• If the pebbles are huge, you can't drive (the sound is trapped).
• If the pebbles are tiny but there are too many of them, you still can't drive smoothly.
• The authors found the "Goldilocks" zone. They proved that even if there are many pebbles, as long as they get smaller at a specific rate (related to that logarithmic rule), the car (the sound wave) will drive smoothly forever, just as if the road were perfectly paved.

Why Does This Matter?

In the world of quantum mechanics (the physics of tiny particles), these "pianos" represent how electrons move through a crystal lattice.

• Existence of Wave Operators: Means the electron can travel through the material without getting stuck. It conducts electricity.
• Completeness: Means every electron in the material can travel freely. There are no "trapped" electrons.

The authors show that for a very wide class of materials (those satisfying the Szegő condition and their new logarithmic rule), the electrons will always flow freely. This helps physicists understand which materials will be good conductors and which might be insulators.

The "Secret Weapon": The Unit Circle Trick

How did they solve this? The math of the piano (Jacobi matrices) is hard. But the authors used a clever magic trick.

They realized that the problem of the piano on a straight line is mathematically identical to a problem involving polynomials on a circle (Unit Circle).

• Imagine wrapping the infinite piano into a giant loop.
• On this loop, the math becomes much easier to handle using tools from "harmonic analysis" (the study of waves and frequencies).

They proved a new lemma about how these waves behave on the circle, specifically looking at how they concentrate in small arcs. This technical proof was the key that unlocked the solution for the piano.

Summary

1. The Problem: Can we predict how waves move through a slightly imperfect, infinite system?
2. The Condition: The imperfections must follow a "Szegő" rule (they aren't too wild).
3. The New Rule: The authors added a specific "logarithmic speed limit" to ensure the imperfections fade fast enough.
4. The Result: If this rule is met, the system behaves exactly like a perfect, free system. The waves travel freely, and nothing gets stuck.
5. The Method: They solved a hard linear problem by turning it into a circular one, using advanced math to prove the waves behave nicely.

In short, Denisov and Young have drawn a new, sharper line in the sand, telling us exactly how much "messiness" a system can have before it stops conducting energy perfectly.

Technical Summary

1. Problem Statement

The paper investigates the scattering theory for semi-infinite bounded self-adjoint Jacobi matrices $J$. Specifically, it addresses the existence and completeness of wave operators $\Omega^\pm$ defined as the strong limits:
$$\Omega^\pm = \lim_{T \to \pm \infty} e^{-iTJ} e^{iTJ_0}$$
where $J_0$ is the "free" Jacobi operator (with coefficients $b_n = 1/2, v_n = 0$).

The central question is: Under what conditions on the spectral measure $\rho$ of $J$ (or equivalently, on the Jacobi coefficients) do these wave operators exist, and do they map the entire Hilbert space $\ell^2(\mathbb{N})$ onto the absolutely continuous subspace $\ell^2_{ac}(J)$?

The authors focus on the Szegő class of measures. A measure $\rho$ supported on $[-1, 1]$ is in the Szegő class if its density $\rho'$ with respect to the equilibrium measure satisfies:
$$\int_{-1}^1 \log \rho'(x) \frac{dx}{\sqrt{1-x^2}} > -\infty$$
While the Szegő condition guarantees that the perturbation $J - J_0$ is Hilbert-Schmidt, it is not sufficient on its own to guarantee the existence of wave operators without additional quantitative decay conditions on the tail of the perturbation.

2. Methodology

The authors employ a sophisticated blend of Inverse Spectral Theory, Orthogonal Polynomials on the Unit Circle (OPUC), and Harmonic Analysis.

• Szegő Mapping and OPUC: Instead of working directly with Orthogonal Polynomials on the Real Line (OPRL), the authors utilize the Szegő mapping to transform the problem to the unit circle $\mathbb{T}$. This maps the measure $\rho$ on $[-1, 1]$ to an even measure $\sigma$ on $\mathbb{T}$. The Jacobi coefficients are related to the Verblunsky coefficients $\{\gamma_n\}$ of $\sigma$ via Geronimus relations.
• Asymptotic Analysis of Polynomials: The core of the proof relies on analyzing the asymptotic behavior of the orthonormal polynomials $\Phi_n^*(z)$ (reversed monic polynomials) associated with $\sigma$. Under the Szegő condition, $\Phi_n^*$ converges to the inverse Szegő function $D^{-1}$ in $L^2_\sigma$.
• Wave Packet Decomposition: The evolution $e^{iTJ_0}$ is analyzed by decomposing initial states into "wave packets" localized in frequency space. The authors use a partition of unity on the circle to localize these packets.
• Non-Stationary Phase: To control the free evolution $e^{iTJ_0}$, the authors apply the method of non-stationary phase to estimate the decay of Fourier coefficients of the evolved wave packets.
• Key Technical Lemma (Theorem 2.7): A crucial step is proving that a specific weighted sum of polynomials converges to the Szegő function limit. This requires establishing that the "error" terms arising from the difference between the perturbed polynomials and the free limit vanish under the specific decay condition imposed on $\gamma_n$.

3. Key Contributions and Results

A. Main Theorem (Theorem 1.1)

The primary result establishes the existence and completeness of wave operators under a specific quantitative condition on the Verblunsky coefficients $\{\gamma_n\}$.

Condition: The authors assume $\{\gamma_n\} \in \ell^2(\mathbb{Z}^+)$ (which is equivalent to the Szegő condition) and impose the additional tail decay condition:
$$\lim_{n \to \infty} \log(n) \sum_{s=n}^{2n} |\gamma_s|^2 = 0$$
Result: Under this condition, the strong limits $\Omega^\pm$ exist, and they are complete, meaning $\text{Ran}(\Omega^\pm) = \ell^2_{ac}(J)$. Consequently, every state in the absolutely continuous subspace behaves asymptotically like a free particle.

B. Corollary on Jacobi Coefficients (Corollary 1.2)

Using the direct Geronimus relations, the authors translate the condition on $\gamma_n$ back to the original Jacobi coefficients $\{a_n\}$ (diagonal) and $\{b_n\}$ (off-diagonal).
They define a class $\delta \ell^2_{\log}$ of sequences satisfying the logarithmic decay condition. The corollary states that if:
$$\{v_n\} \in \delta \ell^2_{\log} + \ell^1(\mathbb{N}) \quad \text{and} \quad \{b_n - 1/2\} \in \delta \ell^2_{\log} + \ell^1(\mathbb{N})$$
then the wave operators exist and are complete.

C. Technical Estimate on Localized Polynomials

In the Appendix, the authors prove a new estimate (Lemma 4.1) on the norm of orthogonal polynomials $\Phi_n$ localized to small arcs on the unit circle. This estimate bounds the $L^2$ mass of the polynomials on an arc $I$ in terms of the length of the arc and the sum of Verblunsky coefficients near the index $n$. This is noted as being of independent interest for OPUC theory.

D. Connection to Pointwise Convergence

The authors discuss the Pointwise Convergence Assumption (PCA), which posits that $\Phi_n^*(z)$ converges pointwise almost everywhere. They show that the PCA implies their main convergence result (Theorem 2.7). They note that while the PCA is known to hold if $\sum |\gamma_n|^2 \log n < \infty$ (Menshov-Rademacher), their condition is strictly weaker, bridging a gap between known sufficient conditions and the optimal Szegő regime.

4. Significance

• Optimality: The result operates in the "transitional" regime. It is known that if the perturbation decays too slowly (e.g., purely singular continuous spectrum), wave operators do not exist. If the perturbation is trace-class, Kato-Rosenblum theory guarantees existence. This paper pushes the boundary significantly closer to the pure Szegő condition (Hilbert-Schmidt perturbations), identifying a precise logarithmic decay threshold that ensures scattering.
• Methodological Innovation: The work demonstrates the power of transferring discrete Schrödinger operator problems to the unit circle via OPUC techniques. The use of wave packet decomposition combined with OPUC asymptotics provides a robust framework for analyzing time-dependent scattering in discrete systems.
• Comparison to Continuous Models: The results parallel recent advances in continuous Dirac and Schrödinger operators, confirming that the Szegő condition is the natural boundary for the existence of wave operators in both discrete and continuous settings, provided a mild tail condition is met.

In summary, Denisov and Young provide a definitive answer to the scattering problem for Jacobi matrices in the Szegő class, proving that a mild logarithmic decay of the perturbation tail is sufficient to ensure that the system scatters completely to the free evolution.