On uniqueness of radial potentials for given Dirichlet spectra with distinct angular momenta

Imagine you are a detective trying to solve a mystery inside a spherical room (like a tiny planet or a star). Inside this room, there is an invisible force field (called a potential) that affects how waves, like sound or light, bounce around.

Your goal is to figure out exactly what this invisible force field looks like, but you can't see it directly. All you have are the echoes (the spectra) of the waves bouncing off the walls.

This paper is about a specific type of detective work: Can you identify the invisible force field just by listening to the echoes of waves spinning at different speeds?

Here is a breakdown of the paper's findings using simple analogies:

1. The Setup: The Spinning Waves

Think of the room as a drum. When you hit a drum, it vibrates in specific patterns. In this 3D room, the waves can also spin.

Angular Momentum ( $\ell$ ): Imagine the waves spinning like a top.
- $\ell=0$ : The wave isn't spinning at all (it just pulses in and out).
- $\ell=1$ : The wave spins once.
- $\ell=2$ : The wave spins twice, and so on.
The Echo (Spectrum): For each spin speed, the room has a unique set of "notes" (frequencies) it can sing. These notes are the Dirichlet spectra.

2. The Big Question

In the past, scientists knew that if you listened to one spinning speed (say, just $\ell=0$ ), you couldn't uniquely identify the force field. It's like trying to guess the shape of a room just by hearing one specific echo; many different rooms could produce that same sound.

The big question this paper asks is: If you listen to the echoes of two different spinning speeds (two different angular momenta), can you finally pinpoint the exact shape of the invisible force field?

3. The Main Discoveries

The "Infinite List" Solution (The Easy Case)

First, the authors proved a global rule: If you have the echo data for infinitely many different spinning speeds (as long as the list of speeds isn't too sparse), you can 100% uniquely determine the force field.

Analogy: If you have a fingerprint scanner that can read your fingerprints from every possible angle, there is no doubt who you are.

The "Two-Note" Solution (The Hard Case)

The real magic of this paper is what happens when you only have data from two specific spinning speeds.

The Conjecture: A previous guess by other scientists (Rundell and Sacks) suggested that any two different spinning speeds should be enough to solve the mystery.
The Proof: The authors proved this guess is true for three specific pairs of speeds:
1. No spin ( $\ell=0$ ) and one spin ( $\ell=1$ ).
2. One spin ( $\ell=1$ ) and two spins ( $\ell=2$ ).
3. No spin ( $\ell=0$ ) and three spins ( $\ell=3$ ).

They showed that near a "flat" force field (where the room is empty), these two sets of echoes are enough to uniquely identify the hidden force.

4. How Did They Do It? (The Detective's Toolkit)

The authors didn't just guess; they used a sophisticated mathematical toolkit:

The Kneser–Sommerfeld Formula: Think of this as a universal translator. It's a complex mathematical identity that connects the echoes of different spinning speeds. It allows the detectives to translate the "language" of a spinning wave into the "language" of a non-spinning wave, revealing hidden connections.
Transformation Operators: Imagine these as magic lenses. The authors used these lenses to take the messy, curved waves (Bessel functions) and turn them into simple, straight waves (trigonometric functions). This made the math much easier to handle.
The "Odd/Even" Trick: They realized that if a hidden force field existed that fooled the system (making two different fields look the same), that "fake" field would have to behave in a very strange, symmetrical way (like a mirror image). By proving that such a symmetrical fake field would have to be zero (nothing), they proved the real field must be unique.

5. The Computer's Role

For the most difficult case ( $\ell=0$ and $\ell=3$ ), the math became so complicated that the equations turned into an 8th-order differential equation (a very complex recipe for change).

The authors used a computer to simulate the solution.
The computer showed that any "fake" solution would explode into infinity at the edges of the room, which is physically impossible for a real wave.
This confirmed that the only valid solution is the correct one.

Summary

This paper is a victory for mathematical physics. It confirms that two different perspectives (two angular momenta) are often enough to see the whole picture.

Before: We needed a lot of data or extra information to find the hidden force.
Now: We know that for specific pairs of spinning speeds, the echoes alone are a unique fingerprint.

It's like realizing that if you listen to a song played on a piano with your eyes closed, and then listen to the same song played on a violin, you can perfectly reconstruct the sheet music, even without seeing the notes. The combination of the two "instruments" (angular momenta) reveals the secret.

Here is a detailed technical summary of the paper "On uniqueness of radial potentials for given Dirichlet spectra with distinct angular momenta" by Gobin, Grébert, Helffer, and Nicoleau.

1. Problem Statement

The paper addresses an inverse spectral problem for radial Schrödinger operators with singular potentials on the unit interval $(0, 1)$ . The operator is defined by the equation:
$-u''(r) + \left( \frac{\ell(\ell+1)}{r^2} + q(r) \right) u(r) = \lambda u(r),$
subject to Dirichlet boundary conditions at $r=1$ and regularity at $r=0$ . Here, $\ell \in \mathbb{N}$ is the angular momentum quantum number, and $q \in L^2(0, 1)$ is the unknown radial potential.

The Core Question: Can the potential $q$ be uniquely determined solely from the knowledge of the Dirichlet spectra (eigenvalues) corresponding to distinct values of angular momentum $\ell$ , without the use of norming constants (Neumann data)?

Context: Classical results (Borg–Levinson, Carlson) require norming constants for uniqueness. While knowing the spectrum for a single $\ell$ yields an infinite-dimensional isospectral set, combining spectra from two distinct $\ell$ 's was conjectured by Rundell and Sacks (2001) to yield uniqueness, though this remained unproven for general pairs.

2. Methodology

The authors employ a multi-faceted analytical approach combining spectral theory, special functions, and differential equations:

Linearization and Spectral Map:
The problem is analyzed in a neighborhood of the zero potential ( $q \equiv 0$ ). The authors define a spectral map $S_{\ell_1, \ell_2}$ mapping the potential to the renormalized eigenvalues. The core strategy is to prove that the Fréchet differential of this map at $q=0$ is injective. If the differential is injective and has a closed range, local uniqueness follows via the inverse function theorem (or a variant for Banach spaces).
Kneser–Sommerfeld Formula:
A central tool is the corrected Kneser–Sommerfeld expansion, a Fourier–Bessel series identity. The authors utilize the corrected version (noting Watson's original omission of an integral term) to derive integral identities satisfied by the difference of two potentials, $\zeta = \tilde{q} - q$ , assuming they share the same spectra.
Transformation Operators:
To handle the Bessel functions arising from the singular term $\ell(\ell+1)/r^2$ , the authors use index-reduction operators ( $S_\ell$ ) and composite operators ( $T_\ell$ ) introduced by Rundell and Sacks. These operators map Bessel kernels to trigonometric kernels, effectively reducing the problem to analyzing functions on the interval with respect to trigonometric bases.
Symmetry and Differential Relations:
By exploiting the reflection symmetry about the midpoint $x=1/2$ (via the map $\sigma(x) = 1-x$ ), the authors transform the integral orthogonality conditions into linear ordinary differential equations (ODEs) with singular coefficients. The function $\zeta$ (or its derivatives) is shown to satisfy high-order ODEs with specific boundary conditions at $x=1/2$ .
Computer-Assisted Analysis:
For the case $(\ell_1, \ell_2) = (0, 3)$ , the resulting differential equation is of 8th order. The authors use numerical integration and Frobenius analysis to demonstrate that any non-trivial solution to the associated ODE system fails to be square-integrable ( $L^2$ ), thereby proving uniqueness by contradiction.

3. Key Contributions and Results

A. Global Uniqueness (Infinitely Many Spectra)

Theorem 1.1: If the Dirichlet spectra are known for an infinite sequence of angular momenta $(\ell_k)_{k \ge 1}$ satisfying the Müntz condition $\sum \frac{1}{\ell_k} = +\infty$ , the potential $q$ is uniquely determined globally. This result connects Dirichlet spectral data to scattering phases.

B. Completeness of Squared Eigenfunctions

Theorem 1.2: Under the same Müntz condition, the family of squared eigenfunctions $\{ \psi_{\ell_k, n}^2 \}_{k,n}$ is complete in $L^2(0, 1)$ . This establishes the density of the linearized spectral data.

C. Local Uniqueness for Two Spectra (Main Result)

The paper proves Theorem 1.3, establishing local uniqueness near the zero potential for specific pairs of angular momenta:
$(\ell_1, \ell_2) \in \{ (0, 1), (1, 2), (0, 3) \}.$
For these pairs, the knowledge of two Dirichlet spectra uniquely determines the potential in an $L^2$ -neighborhood of zero.

Case $(0, 1)$ : The differential of the spectral map is an isomorphism (injective and surjective).
Cases $(1, 2)$ and $(0, 3)$ : The differential is injective with a closed range. Local injectivity is established via the Open Mapping Theorem and mean value arguments.
Case $(0, 2)$ : The authors prove the differential is injective but leave the surjectivity (and thus the full local uniqueness via the inverse function theorem) as an open conjecture.

D. Technical Breakthroughs

Correction of Kneser–Sommerfeld: The paper rigorously applies the corrected integral identity, which is essential for deriving the moment conditions.
High-Order ODE Analysis: The derivation of the 4th-order ODE for $(0, 2)$ and the 8th-order ODE for $(0, 3)$ , combined with symmetry arguments, allows the reduction of the inverse problem to a uniqueness problem for ODEs.
Frobenius Analysis: For the $(0, 3)$ case, the authors identify complex indicial roots ( $\rho = -2 \pm 2i\sqrt{11}$ ) leading to oscillatory blow-up at the boundaries, proving that no non-trivial $L^2$ solution exists.

4. Significance

Confirmation of Conjecture: The results confirm the Rundell–Sacks conjecture (2001) for specific configurations, demonstrating that angular momentum channels encode distinct spectral information sufficient to resolve the potential without norming constants.
Refinement of Carlson–Shubin: The work sharpens previous results by Carlson and Shubin (1994), which required the difference $\ell_2 - \ell_1$ to be odd for finite-dimensional isospectral sets. This paper shows injectivity even when the difference is even (e.g., $(0, 2)$ and $(0, 3)$ ).
Methodological Advancement: The combination of transformation operators with the Kneser–Sommerfeld formula provides a robust framework for analyzing singular Sturm–Liouville problems. The use of computer-assisted proofs for high-order differential equations opens a pathway for tackling more complex pairs $(\ell_1, \ell_2)$ .
Physical Relevance: The results are directly applicable to quantum scattering in bounded domains and the study of acoustic modes in stellar interiors, where only spectral data (frequencies) are observable, not the full wavefunction derivatives.

Conclusion

The paper successfully establishes that for radial Schrödinger operators, the potential is locally uniquely determined by the Dirichlet spectra of two distinct angular momenta in the cases $(0,1)$ , $(1,2)$ , and $(0,3)$ . The authors conjecture that this holds for any distinct pair $(\ell_1, \ell_2)$ , suggesting that the "spectral gap" created by distinct centrifugal terms is sufficient to solve the inverse problem.