🔢 mathematics

Average lower bound on magnetic double-well tunneling with asymptotically large well separation

This paper establishes lower bounds on the magnetic tunneling splitting and hopping coefficient for two identical, non-symmetric potential wells at asymptotically large separation in a constant magnetic field, showing that these quantities decay no faster than $e^{-b\lambda d_1^2-Cd_1^{1+\delta}}$ for almost all separations.

Original authors: Charles L. Fefferman, Say-Yeon Kwon, Jacob Shapiro, Michael I. Weinstein

Published 2026-10-20

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Charles L. Fefferman, Say-Yeon Kwon, Jacob Shapiro, Michael I. Weinstein

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

In the quantum world, particles do not always stay put. Even when trapped in a deep valley of energy, a particle can sometimes vanish from one side and reappear on the other, passing through a barrier that should be impossible to cross. This phenomenon, known as tunneling, is the engine behind many modern technologies, from the flash memory in our phones to the way stars burn. When two such valleys, or "wells," sit near each other, the particle can hop between them, creating a delicate link between the two locations. The speed of this hop depends on how far apart the wells are and how strong the forces holding the particle in place are. Scientists have long understood how this works when no magnetic field is present, but the introduction of a magnetic field changes the rules entirely. The field acts like a subtle, invisible hand that twists the particle's path, turning a simple hop into a complex dance of phases that can either help the particle cross or cancel its movement entirely.

A team of mathematicians at Princeton University and the University of Vienna has now mapped out exactly how this magnetic influence behaves when the two wells are pulled very far apart. They studied a system where two identical energy traps are separated by a vast distance, while a constant magnetic field pushes against them. Their goal was to determine if the particle could still tunnel between these distant points, or if the magnetic field would eventually silence the connection completely. In many cases, when waves interact, they can cancel each other out, leading to a total loss of signal. The researchers wanted to know if this cancellation could happen here, effectively stopping the tunneling, or if a connection would always remain, however faint.

The team proved that for a sufficiently weak magnetic field, the connection between the two wells never truly disappears, even as the distance grows. They showed that while the probability of the particle hopping drops dramatically as the wells move apart, it does not vanish to zero. Instead, it follows a predictable pattern of decay, shrinking at a rate determined by the square of the distance between the wells. Crucially, they demonstrated that this connection holds true for almost every possible distance, with only a tiny, negligible set of exceptions where the signal might momentarily fade. This finding is significant because it confirms that the magnetic field, while it complicates the journey, does not destroy the fundamental ability of the quantum particle to communicate across the gap.

To reach this conclusion, the researchers had to navigate a landscape where standard tools often fail. In the absence of a magnetic field, the mathematics of tunneling is relatively straightforward because the particle's behavior is positive and predictable. However, the magnetic field introduces a swirling, oscillating factor that can cause the tunneling signal to flip between positive and negative values. If these flips align just right, they could theoretically cancel the signal out completely. The team had to prove that such perfect cancellation is not the norm. They developed a new method that combined a precise description of how magnetic fields affect particle motion with a sophisticated technique for averaging the results over a range of distances. This approach allowed them to look past the rare, confusing moments where the signal might dip and see the steady, underlying truth that the connection persists.

The study focused on a specific scenario where the strength of the magnetic field and the depth of the energy wells were fixed, while the distance between the wells was allowed to grow infinitely large. This setup mimics the conditions found in certain crystalline materials, where atoms are arranged in a grid and electrons must hop from one atom to another. The researchers found that as the distance increases, the tunneling probability decreases exponentially, but the rate of this decrease is governed by a specific mathematical relationship involving the magnetic field strength and the distance. They proved that for almost all distances, the tunneling rate stays above a certain threshold, ensuring that the two wells remain linked. This result rules out the possibility that the magnetic field could completely sever the bond between the wells in a typical setup, providing a rigorous foundation for understanding how electrons move in magnetic materials.

One of the most important aspects of this work is that it does not rely on the wells having a perfect, symmetrical shape. In many previous studies, scientists assumed the energy traps were perfectly round or mirrored to make the math easier. This new research removes that requirement, showing that the result holds true even if the wells are irregular or lopsided. This makes the finding much more applicable to real-world materials, which are rarely perfect. The team also clarified that while the magnetic field can make the tunneling slower, it does not change the fundamental scaling law of how the probability drops with distance. The connection weakens, but it does so in a way that is predictable and robust, rather than chaotic or prone to total failure.

The researchers' proof relies on a deep understanding of how complex numbers behave in this physical context. They treated the distance between the wells not just as a physical measurement, but as a variable that could be extended into the realm of complex numbers to reveal hidden patterns. By doing so, they were able to show that the tunneling coefficient, which measures the strength of the link, cannot stay at zero for any significant stretch of distance. They found that while there are specific, isolated distances where the signal might be weaker, these points are so sparse that they do not affect the overall behavior of the system. In essence, the magnetic field adds a layer of complexity, but it does not break the bridge between the two worlds.

This work provides a clear answer to a question that had lingered in the field of quantum physics: does a magnetic field eventually stop a particle from tunneling between distant wells? The answer is no. The connection remains, governed by a precise mathematical law that holds true for the vast majority of distances. This insight helps physicists and engineers better predict how quantum materials will behave under magnetic influence, which is essential for developing new technologies that rely on the controlled movement of electrons. The study confirms that even in the presence of a twisting magnetic force, the quantum world maintains its ability to connect distant points, ensuring that the flow of information and energy across a material is never completely cut off.

Technical Summary: Average Lower Bound on Magnetic Double-Well Tunneling

Problem Statement
This paper investigates the quantum tunneling phenomenon between two identical, smooth, compactly supported potential wells in the presence of a constant perpendicular magnetic field. The study focuses on the regime where the coupling strength and magnetic field are fixed, while the separation distance ( $2d_1$ ) between the well centers tends to infinity.

The central challenge addressed is establishing rigorous lower bounds on the hopping coefficient ( $\rho$ ) and the resulting eigenvalue splitting ( $\Delta$ ) of the two lowest energy levels. In the absence of a magnetic field, semiclassical analysis and the positivity of the ground state allow for standard lower bounds. However, the introduction of a magnetic field introduces complex phases via magnetic translations, leading to an oscillatory hopping integral. While decay estimates provide upper bounds, they do not preclude destructive interference (cancellations) that could cause the hopping coefficient to vanish or become exponentially smaller than the expected Gaussian decay rate. Previous work by the authors and others has shown that for certain non-radial, non-positive potentials, exact degeneracy (vanishing tunneling) can occur. This paper aims to prove that, for sufficiently small magnetic fields, such vanishing does not occur generically; specifically, it establishes that the tunneling remains significant outside a set of separations of density zero.

Methodology
The proof strategy combines explicit resolvent kernel formulas with complex analytic techniques, specifically logarithmic averaging. The methodology proceeds through the following steps:

Integral Representation: The authors utilize the Landau resolvent kernel to express the hopping coefficient $\rho(d_1)$ as a double integral over the compact support of the potential. This representation involves Tricomi's confluent hypergeometric function, $U(\nu, 1, z)$ .
Holomorphic Continuation: Unlike previous approaches that complexified the coupling parameter (which altered the single-well eigenvalue), this work fixes the magnetic field and coupling. Instead, the separation distance $d_1$ is treated as a complex variable $z$ . The authors demonstrate that the integrand remains holomorphic in the half-plane $\text{Re}(z) > a$ (where $a$ is the support radius of the potential) because the argument of the Tricomi function avoids its branch cut.
Gaussian Quotient and Sector Bounds: To isolate the dominant Gaussian decay $e^{-B d_1^2}$ (where $B = b\lambda$ ), the authors define a quotient function $q(z) = e^{Bz^2}\rho(z)$ . They establish a global upper bound for $|q(z)|$ on a translated sector in the complex plane, showing it grows at most exponentially with a linear exponent.
Non-Vanishing at a Point: Using continuity arguments from the non-magnetic case ( $b=0$ ), where the ground state is strictly positive, the authors prove that $\rho(d_1)$ is non-zero for a specific fixed separation when the magnetic field $b$ is sufficiently small.
Complex Analysis Lemma (Logarithmic Averaging): The core of the proof applies a lemma derived from Blaschke factorization. By bounding the function on a sector and knowing it is non-zero at one point, the lemma provides an estimate for the logarithmic average of the function along the real axis. This allows the authors to bound the measure of the set where the function is "too small."

Key Contributions and Results

Theorem 1.1 (Lower Bound on Hopping): For a sufficiently small magnetic field strength $b$ , and for any fixed $\delta \in (0, 1)$ , there exists a set of separations $Z_\delta$ of density zero such that for all $d_1 \notin Z_\delta$ :
$|\rho(d_1)| \geq e^{-b\lambda d_1^2 - C d_1^{1+\delta}}$
This result confirms that the leading exponential decay is exactly determined by the magnetic field and separation ( $e^{-B d_1^2}$ ), with only a sub-quadratic correction.
Theorem 1.2 (Lower Bound on Eigenvalue Splitting): As a consequence of the hopping bound and a two-state reduction, the eigenvalue splitting $\Delta(d_1)$ satisfies the same lower bound outside the density-zero set:
$\Delta(d_1) \geq e^{-b\lambda d_1^2 - C d_1^{1+\delta}}$
Genericity of Tunneling: The paper establishes that while exact degeneracy (vanishing tunneling) is possible for specific configurations (as shown in prior counterexamples), it is not generic. The set of separations where the tunneling coefficient drops below the derived lower bound has Lebesgue density zero.
Relaxation of Symmetry Assumptions: Unlike earlier results that required radial or reflection symmetry of the potential to ensure a definite sign for the hopping coefficient, this proof holds for general smooth, compactly supported potentials without assuming symmetry or non-degeneracy of the potential minimum.

Significance and Claims
The authors claim that this work resolves the precise Gaussian distance scale for magnetic tunneling in the large-separation limit. The significance lies in:

Rigorous Justification of Tight-Binding Models: The results provide the necessary lower bounds on the magnetic hopping coefficient required for the rigorous derivation of discrete effective (tight-binding) magnetic Hamiltonians for macroscopic crystalline quantum materials. Without such bounds, the limiting models could fail for certain atomic configurations.
Comparison of Decoupling Mechanisms: The paper distinguishes between two regimes of weakening interaction: increasing the coupling strength (deep wells) versus increasing the separation distance. This work specifically addresses the large-separation regime, showing that the leading exponent remains $B d_1^2$ with a sub-quadratic correction, contrasting with previous bounds that required different scaling in the exponent.
Analytic Technique: The method of complexifying the separation distance while keeping the single-well problem fixed allows for the use of the full Landau resolvent without the need for cutoff Hamiltonians or analytic continuation of the potential itself, simplifying the analysis compared to previous works on this topic.

The paper concludes that for generic large separations, the magnetic field does not eliminate tunneling but merely modulates it with an oscillatory factor that is bounded away from zero on a set of full density.

Technical Summary: Average Lower Bound on Magnetic Double-Well Tunneling

More like this