🔢 mathematics

Unique continuation at infinity for potentials with arbitrary radial growth

This paper establishes a Landis-type unique continuation theorem at infinity for solutions to the Schrödinger equation $\Delta u = Vu$ with radially growing potentials, proving that solutions decaying faster than an explicitly computable threshold $e^{-g(r)}$ must vanish identically.

Original authors: Henrik Ueberschaer

Published 2026-08-10

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Henrik Ueberschaer

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 are standing in a vast, invisible ocean where the water isn't made of H₂O, but of mathematical possibilities. This is the world of partial differential equations, a branch of science that describes how things change and move through space. Think of it as the ultimate rulebook for how heat spreads, how sound waves travel, or how a quantum particle dances. In this ocean, there are "potentials"—invisible hills and valleys in the landscape that push or pull on whatever is moving through them.

The big question this paper tackles is about "unique continuation at infinity." In plain English, it asks: If you have a wave (or a particle) moving through this ocean, and it gets smaller and smaller as it travels further and further away, can it eventually vanish completely? Or does the math force it to keep a tiny, stubborn spark of life, no matter how far out you go? For decades, mathematicians have wondered if there is a specific speed limit for how fast these waves can fade away. If they fade too quickly, the math says they must have been zero to begin with. But if they fade just a little slower, they might survive. This paper explores exactly where that line is drawn, especially when the "hills and valleys" of the ocean get steeper and steeper as you go further out.

The Story of the Stubborn Wave

In this paper, author Henrik Ueberschär dives into a specific corner of this mathematical ocean where the landscape is perfectly symmetrical—like a series of concentric rings stretching out to infinity. He looks at a real-valued wave (think of it as a simple, physical ripple, not a complex, ghostly one) moving through a landscape defined by a potential $V$ . The rule of the game is that the wave follows the equation $\Delta u = V u$ , which is a fancy way of saying the wave's shape is determined by the terrain it's traveling over.

The central mystery is: How fast can this wave disappear?

Imagine you are walking away from a campfire. If the wind (the potential) is gentle, the smoke (the wave) might drift away and vanish quickly. But what if the wind gets stronger and stronger the further you go? Does the smoke vanish instantly, or does it cling to the air? Ueberschär proves that no matter how wild and fast the wind grows, the smoke cannot vanish too quickly. There is a "decay threshold"—a minimum speed at which the wave must fade. If it tries to fade faster than this limit, it's a mathematical impossibility; the wave must have been zero to begin with.

The Main Discovery: The "Speed Limit" of Fading

The paper's main finding is the construction of a specific "decay threshold" for these waves. Ueberschär shows that for any continuous, positive function $G$ that describes how fast the potential grows, there is a corresponding function $\beta(r)$ that acts as a speed limit.

Here is the punchline: If a real-valued solution (a real wave) starts somewhere (specifically, if it's not zero at the very center), it cannot fade away faster than $e^{-\beta(r)}$ along a sequence of points stretching out to infinity.

Think of it like a runner trying to outrun a shadow. The shadow represents the "decay threshold." No matter how fast the runner (the wave) tries to fade, the shadow (the math) says, "You can't go faster than this." The paper proves that there will always be a sequence of points, getting further and further away, where the wave is still at least as big as $c \cdot e^{-\beta(r)}$ . It's a guarantee that the wave leaves a trail of breadcrumbs that never completely disappears.

Two Different Scenarios

The author provides two main ways to calculate this speed limit, depending on how much we know about the terrain:

The General Case (The "Any Shape" Rule):
If we only know that the potential grows according to some continuous function $G$ (it could be bumpy, wiggly, or smooth), the paper defines a function $\beta(r)$ that includes the dimension of space ( $n$ ) and the total "area" under the growth curve of $G$ .
- The Result: The wave must satisfy $|u(x)| \ge c e^{-\beta(|x|)}$ at some points far away.
- The Analogy: Imagine you are walking through a forest where the trees get taller and taller. Even if you don't know exactly how the trees are arranged, as long as you know the general trend of their height, you know you can't disappear into the fog faster than a certain rate. The paper gives you the formula for that rate.
The Smooth Case (The "Agmon Distance" Connection):
If the potential is very smooth (twice differentiable) and grows in a predictable way, the paper finds a sharper, more precise threshold. Here, the decay rate is linked to something called the "Agmon distance," which is a concept from physics used to describe how quantum particles tunnel through barriers.
- The Result: The wave must satisfy $|u(x)| \ge c e^{-\beta g(|x|)}$ , where $g(r)$ is the integral of the square root of the potential's growth.
- The Analogy: This is like knowing the exact topography of a mountain range. Instead of just guessing the fog's speed, you can calculate the exact "tunneling" distance the wave has to travel. The paper shows that for smooth, growing potentials, the wave's survival rate is directly tied to this geometric distance.

What the Paper Rules Out

It is crucial to understand what this paper says doesn't happen. The paper explicitly argues against the idea that a real-valued wave can vanish super-fast, even if the potential grows wildly.

No "Super-Fast" Vanishing: You might think that if the potential $V$ grows infinitely fast, it could crush the wave into nothingness instantly. The paper proves this is false. Even with arbitrary radial growth, the wave cannot decay faster than the calculated threshold.
Real vs. Complex: The paper emphasizes that this "stubbornness" only applies to real-valued solutions. If the wave were complex-valued (a concept involving imaginary numbers, common in quantum mechanics), the rules are different, and the wave can fade faster. But for real, physical-looking waves, the math is strict: they must leave a trace.

Why This Matters

This isn't just about abstract math; it's about understanding the fundamental limits of how things behave in our universe. The paper answers a long-standing question about the "Landis conjecture," which asked if there is a universal limit to how fast a wave can fade.

By proving that a "decay threshold" exists for potentials with arbitrary radial growth, Ueberschär shows that nature has a safety net. No matter how extreme the environment gets (as long as it's radial and continuous), a real wave cannot simply vanish into thin air without a trace. It must maintain a minimum presence, however small, as it travels toward infinity.

The paper doesn't just suggest this; it proves it using rigorous mathematical arguments involving contradiction. The author assumes the wave does vanish too fast, follow the math through a series of logical steps involving integrals and derivatives, and eventually hit a contradiction (like proving $n+1 \le n$ ). This contradiction confirms that the initial assumption was wrong, and the wave must indeed survive.

In the end, the paper gives us a new set of tools to measure the "stubbornness" of waves. Whether the potential grows like a polynomial (a gentle curve) or something much wilder, we now have a formula to calculate exactly how much of the wave must remain. It's a reminder that in the vast, expanding universe of mathematics, some things are simply too persistent to disappear completely.

Technical Summary: Unique Continuation at Infinity for Potentials with Arbitrary Radial Growth

Problem Statement
This paper addresses the problem of unique continuation at infinity for real-valued solutions to the Schrödinger equation $\Delta u = V u$ on $\mathbb{R}^n$ . Specifically, it investigates the decay properties of solutions when the potential $V$ is radial and may grow rapidly as $|x| \to \infty$ . The work extends the scope of Landis' conjecture, which originally posited that solutions decaying faster than $e^{-c|x|}$ must be trivial for bounded potentials. While Meshkov established that complex-valued solutions can decay as slowly as $e^{-c|x|^{4/3}}$ , and recent work by Frank and Ivanisvili showed the conjecture fails for general real-valued potentials in dimensions $n \ge 3$ , the qualitative version of the conjecture holds for radial potentials. This paper seeks to quantify the decay threshold for radial potentials by establishing explicit lower bounds for non-trivial solutions along a sequence of points, determining how these bounds depend on the growth of $|V|$ .

Methodology
The proofs rely on a contradiction argument involving the spherical mean of the solution. Let $u(r, \theta)$ be the solution in spherical coordinates. The author defines the spherical mean $w(r) = \frac{1}{\text{vol}(S^{n-1})} \int_{S^{n-1}} u(r, \theta) d\theta$ . Since $V$ is radial, $w(r)$ satisfies a second-order ordinary differential equation (ODE):
$w''(r) = (1-n)r^{-1}w'(r) + V(r)w(r).$

The core strategy assumes that the solution decays faster than a specific threshold function $e^{-\beta(r)}$ (or $e^{-\beta g(r)}$ in the refined case). Under this assumption, the function $h(r) = |w(r)|e^{\beta(r)}$ (or $|v(r)|e^{\beta r}$ in the transformed variable) is assumed to tend to zero at infinity. The proof proceeds by:

Establishing that $w$ (or a transformed version $v$ ) does not vanish identically due to the condition $u(0) \neq 0$ and unique continuation principles; the real-valuedness of $u$ is essential for these proofs.
Identifying a maximum point $r_0$ for the weighted function $h(r)$ .
Deriving lower bounds for $|w(r_0)|$ and $|w'(r_0)|$ based on the properties of the maximum.
Integrating the ODE against a smooth cutoff function $\chi$ on the interval $[r_0, T]$ and letting $T \to \infty$ .
Comparing the resulting integral inequalities with the assumed decay bounds to derive a contradiction (typically showing that a derivative bound implies an inequality like $n+1 \le n$ ).

For potentials with specific regularity ( $G \in C^2$ ), the author employs a change of variables $v(r) = w(f(r))$ where $f = g^{-1}$ and $g(r) = \int_0^r \sqrt{G(s)} ds$ . This transformation aligns the decay exponent with the Agmon distance associated with the potential growth.

Key Contributions and Results

Theorem 1.1 (General Growth): For any continuous, non-negative function $G$ such that $|V(x)| \le G(|x|)$ , the paper constructs a decay threshold $\beta(r) = (n+1)r + \int_0^r G(s) ds$ . It proves that if $u$ is a real-valued solution with $u(0) \neq 0$ , there exists a sequence of points $x_i \to \infty$ such that $|u(x_i)| \ge c e^{-\beta(|x_i|)}$ . This result requires no assumptions on the monotonicity or variation of $G$ , though $G$ must be continuous and non-negative.
Corollary 1.2 (Bounded Potentials): In the case where $|V| \le 1$ , the theorem yields a lower bound of order $e^{-(n+2)|x|}$ along a sequence. This provides an explicit upper bound for the quantity $A_u = \liminf_{|x|\to\infty} \frac{\log |u(x)|^{-1}}{|x|}$ , showing $A_u \le n+2$ .
Corollary 1.3 (Arbitrary Growth Construction): The paper demonstrates that for any strictly increasing function $f$ with summable inverse values, one can construct a radial potential growing as $f(n)$ on circles $|x|=n$ such that solutions still satisfy an exponential lower bound along a sequence. This highlights that the decay threshold is robust even for potentials with very rapid growth, provided the growth is concentrated on specific scales.
Theorem 1.4 (Refined Growth with Agmon Distance): Under stronger assumptions ( $G \in C^2$ , $G \ge c > 0$ , and bounded derivatives of $g^{-1}$ ), the decay threshold is refined to $e^{-\beta g(r)}$ , where $g(r) = \int_0^r \sqrt{G(s)} ds$ . The exponent function is closely related to the Agmon distance associated with the potential, and the theorem yields exactly the type of decay appearing in sharp cutoffs for polynomial growth potentials ( $G(r) = r^N$ ).

Significance
The paper establishes a quantitative Landis-type theorem for radial potentials with arbitrary growth rates. Its primary significance lies in:

Generalization: Extending previous results (which were limited to bounded or polynomially growing potentials) to potentials with arbitrary continuous growth, while providing explicit lower bounds for non-trivial solutions along a sequence of points rather than proving global triviality for all faster decays.
Explicit Thresholds: Providing an explicit formula for the decay threshold in terms of the potential's growth function $G$ , distinguishing between a general continuous case and a refined case involving the Agmon metric.
Real-Valuedness: Reinforcing the distinction between real and complex solutions, as the proofs rely fundamentally on the real-valuedness of $u$ to establish the necessary inequalities.
Connection to Agmon Theory: Demonstrating that for sufficiently regular potentials (specifically $C^2$ and bounded away from zero), the unique continuation threshold at infinity is governed by the Agmon distance, linking the problem to semiclassical analysis and eigenfunction decay estimates.

The work does not propose new experimental applications or future directions but serves as a rigorous analytical extension of existing conjectures regarding the non-triviality of solutions to the Schrödinger equation under specific decay conditions.

The Story of the Stubborn Wave

The Main Discovery: The "Speed Limit" of Fading

Two Different Scenarios

What the Paper Rules Out

Why This Matters

More like this