The continuous spectrum of bound states in expulsive potentials

Contrary to common intuition, this paper demonstrates that one- and two-dimensional Schrödinger equations with expulsive potentials steeper than quadratic support a continuous spectrum of normalizable bound states, revealing a phenomenon of effective self-trapping in linear systems that extends to vortex states and stable nonlinear extensions.

The Gist

The Big Surprise: Trapping Things in a "Repelling" Force

Usually, in physics, we think of a "trap" like a bowl. If you put a marble in a bowl, it rolls to the bottom and stays there. This is how atoms are usually held in place in experiments.

But what if you have a hill instead of a bowl? If you put a marble on a steep hill, it doesn't stay; it rolls down and flies away. In quantum physics, this "hill" is called an expulsive potential. Common sense tells us that if you put a quantum particle (like an electron or an atom) on a steep, repelling hill, it should spread out and disappear into the distance. It should be "delocalized."

The paper's main discovery is that this common sense is wrong.

The researchers found that if the hill is steep enough (steeper than a standard parabola), the particle doesn't fly away. Instead, it gets "self-trapped." It stays in a specific, localized spot, even though the force is pushing it away. It's as if you placed a marble on a hill, and instead of rolling off, it started vibrating so intensely in one spot that it effectively pinned itself there.

The "Speeding Car" Analogy

To understand why this happens, imagine a car driving down a very steep, curved hill.

• The Hill: The repulsive force pushing the particle away.
• The Car: The quantum particle.

If the hill is gentle, the car rolls down slowly. But if the hill gets steeper and steeper, the car accelerates incredibly fast.

In the quantum world, speed and "wiggling" (oscillation) are linked. Because the particle is being pushed so hard by the steep hill, it starts "wiggling" or oscillating its wave pattern at a frantic speed. These rapid, chaotic wiggles cancel each other out in the distance, effectively trapping the particle in a small, neat package near the center. The steeper the hill, the tighter the trap.

The Two Main Findings

The paper looked at this in two dimensions (flat surfaces) and one dimension (lines).

1. The "Infinite Spectrum" of Traps
Usually, when we trap something, we only get a few specific "allowed" states (like specific rungs on a ladder). But here, the researchers found that every single energy level works.

• The Analogy: Imagine a piano. Usually, only certain keys make a sound that stays in tune. Here, they found that every key on the piano, from the lowest to the highest, produces a stable, trapped note. This creates a "continuous spectrum" of trapped states.

2. The Vortex (The Swirl)
In the 2D version, they looked at particles that spin or swirl (like a tornado).

• The Analogy: Imagine a whirlpool in a bathtub. Usually, a whirlpool in a repelling force would just fly apart. But they found that if the "hill" is steep enough, you can have a stable, swirling whirlpool that stays in place. They even found exact mathematical formulas for these swirling states.

What About the "Linear" vs. "Nonlinear" Part?

The paper focuses mostly on linear systems.

• Linear (The Main Discovery): This is the "magic" part. The self-trapping happens without the particle interacting with itself. It's purely a result of the shape of the hill. This is surprising because usually, you need particles to interact with each other (nonlinearity) to create a trap.
• Nonlinear (The Side Note): They also briefly checked what happens if the particles do interact (like in a Bose-Einstein Condensate, a super-cold cloud of atoms). They found that the trap still works, but the shape of the trapped particle gets slightly squished or stretched. If the attraction is too strong, the trap can become unstable and the particle might break its symmetry (like a spinning top wobbling and falling over).

Summary of the "Weirdness"

• The Intuition: Steep repulsive forces = particles fly away.
• The Reality: Steep enough repulsive forces = particles get stuck in place due to rapid oscillations.
• The Result: A whole new family of "Bound States in the Continuum." These are particles that are trapped (bound) even though they exist in a range of energies where they should be free (continuum).

Why Does This Matter? (According to the Paper)

The paper suggests this extends our understanding of quantum mechanics and optics (light).

• Optics: Since light waves follow similar math to these particles, this could mean we can trap light in specific ways using special lenses or materials that act like these "steep hills," without needing complex nonlinear materials.
• Quantum Mechanics: It challenges the old rule that you need a "bowl" to trap a particle. You can use a "hill" if it's steep enough.

Note: The paper does not claim this will lead to new medical treatments or specific commercial devices right now. It is a fundamental discovery about how waves behave in extreme environments, offering new theoretical tools for physicists and optical engineers.

Technical Summary

1. Problem Statement

In classical quantum mechanics and Sturm-Liouville theory, there is a fundamental separation between discrete spectra (bound states) and continuous spectra (delocalized states). Intuitively, an expulsive potential (a potential that pushes particles away from the center, such as an inverted harmonic oscillator) should cause quantum states to become widely delocalized, preventing the existence of normalizable bound states.

While "Bound States in the Continuum" (BIC) are known to exist in specific, engineered scenarios (e.g., the von Neumann-Wigner potential or coupled linear systems), they are typically rare, require fine-tuning, or rely on nonlinearity (solitons) for self-trapping. The central problem addressed by this work is whether linear Schrödinger equations with steep expulsive potentials (steeper than the standard quadratic anti-harmonic oscillator) can naturally support a full continuous spectrum of normalizable bound states without the need for nonlinearity or complex coupling.

2. Methodology

The authors employ a combination of analytical asymptotic analysis and numerical simulations to investigate 1D and 2D Schrödinger equations (and their nonlinear Gross-Pitaevskii extensions).

• Governing Equations:

  • 1D Linear Case: The stationary equation is $E\phi = -\frac{1}{2}\phi'' - \frac{1}{2}x^{2\gamma}\phi$, where $\gamma$ determines the steepness of the expulsive potential ($V(x) \propto -x^{2\gamma}$).
  • 2D Linear Case: The equation is expressed in polar coordinates $(r, \theta)$ with integer vorticity $S$ (angular momentum).
  • Nonlinear Extension: The study briefly considers the Gross-Pitaevskii equation (GPE) with cubic nonlinearity ($g|\psi|^2\psi$) to test stability.

• Analytical Approach:

  • The authors derive asymptotic expressions for wave functions at large distances ($|x| \to \infty$ or $r \to \infty$).
  • They utilize WKB-like approximations to determine the behavior of the wave function tails, expanding them in powers of $|x|^{-1}$.
  • They analyze the convergence of the normalization integral $N = \int |\phi|^2 dx$ to determine if states are normalizable (bound) or divergent (delocalized).

• Numerical Approach:

  • Linear: The stationary ordinary differential equations (ODEs) are solved using the Runge-Kutta method with specific initial conditions to find even (fundamental) and odd (dipole) eigenstates.
  • Nonlinear Stability: Perturbed evolution simulations are performed using the split-step Fourier method to test the stability of nonlinear bound states against symmetry-breaking perturbations.

3. Key Contributions and Results

A. Discovery of Linear Self-Trapping in Steep Potentials ($\gamma > 1$)

The most significant finding is counter-intuitive: Expulsive potentials steeper than quadratic ($\gamma > 1$) generate a full continuous spectrum of normalizable bound states.

• Mechanism: As a classical particle rolls down a steep potential hill, it accelerates rapidly. In the quantum counterpart, this rapid acceleration induces a rapidly oscillating phase in the wave function. This high-frequency oscillation causes the wave function to decay effectively, leading to linear self-trapping (localization) without any nonlinearity.
• 1D Results:
  • For $\gamma > 1$, the asymptotic wave function behaves as $\phi(x) \sim |x|^{-\gamma/2} \cos(\frac{|x|^{\gamma+1}}{\gamma+1})$.
  • The norm $N = \int \phi^2 dx$ converges for all eigenvalues $E \in (-\infty, +\infty)$.
  • Both spatially even (fundamental) and odd (dipole) states exist and are normalizable.
• 2D Results:
  • Similar convergence occurs for $\gamma > 1$.
  • The states can carry any integer vorticity $S$ (angular momentum).
  • The asymptotic form includes the vorticity in higher-order corrections.

B. The Critical Case of Quadratic Potential ($\gamma = 1$)

• For the standard inverted harmonic oscillator ($\gamma = 1$), the asymptotic solution involves a logarithmic phase correction: $\phi(x) \sim |x|^{-1/2} \cos(\frac{x^2}{2} + E \ln|x|)$.
• In this case, the normalization integral diverges logarithmically. Thus, strictly speaking, these are not normalizable bound states, distinguishing them from the $\gamma > 1$ case.

C. Exact Solutions for Vortex States

• In the 2D linear case, the authors derived exact analytical solutions for vortex states under a specific constraint: if the potential steepness is tuned to $\gamma = 2S - 1$ (where $S$ is the vorticity), the wave function takes the exact form $\phi(r) \propto r^S \sin(\frac{r^{2S}}{2S})$.
• For $S \ge 2$, these exact solutions are normalizable. For $S=1$ (which implies $\gamma=1$), the norm diverges.

D. Nonlinear Stability (1D)

• The authors investigated the stability of these states when a cubic nonlinearity (self-focusing, $g=-1$) is added.
• Stability Threshold: The linear bound states remain stable under small perturbations if the norm $N$ is below a critical value.
• Parity-Breaking Instability: If the norm exceeds a critical threshold (corresponding to eigenvalues $E \approx -1$ in their specific parameters), the spatially even eigenstate becomes unstable. The symmetry breaks, and the wave packet oscillates or shifts position, demonstrating a "parity-breaking instability."
• Self-defocusing nonlinearity ($g=+1$) was found to deform the states but maintain stability.

4. Significance and Implications

1. Extension of BIC Concept: This work significantly broadens the concept of Bound States in the Continuum (BIC). Traditionally, BICs were thought to require fine-tuned potentials or specific symmetries. This paper shows that a broad class of "steep" expulsive potentials naturally supports a continuous spectrum of bound states.
2. Linear Self-Trapping: It challenges the dogma that self-trapping (localization) is exclusively a nonlinear phenomenon (solitons). It demonstrates that linear systems can exhibit effective self-trapping purely due to the geometry of the potential (steepness) causing rapid phase oscillations.
3. Experimental Relevance:
   • Cold Atoms: These potentials can be realized using blue-detuned optical beams or programmable optical potentials in Bose-Einstein Condensates (BECs).
   • Photonics: Since paraxial optical propagation is mathematically equivalent to the Schrödinger equation, these findings suggest new ways to design optical waveguides and photonic structures that support localized modes in the continuous spectrum, potentially useful for high-Q resonators and lasers.
4. Theoretical Insight: The results provide a deeper understanding of quantum anomalies and the behavior of particles in singular or steep potentials, bridging the gap between classical acceleration and quantum localization.

Conclusion

The paper establishes that steep expulsive potentials ($\gamma > 1$) in 1D and 2D linear Schrödinger equations create a continuous spectrum of normalizable bound states. This "linear self-trapping" is driven by the rapid phase oscillations induced by the potential's steepness. The findings extend the BIC paradigm to a wider class of potentials and offer new avenues for controlling matter waves in cold atoms and light in photonic systems.