⚛️ quantum physics

Time-Dependent Tunneling in the Thin-Barrier Limit

This paper presents a perturbative analysis of time-dependent quantum tunneling through tall, thin barriers where the action is small, revealing that the tunneling probability grows quadratically with time in resonant situations and linearly in non-resonant ones.

Original authors: Tanmay Vachaspati, Frank Wilczek, Zara Yu

Published 2026-08-26

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Tanmay Vachaspati, Frank Wilczek, Zara Yu

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 behave like solid marbles rolling over hills. Instead, they act more like waves of probability, able to slip through barriers that would be completely impenetrable to anything made of ordinary matter. This phenomenon, known as tunneling, is a cornerstone of modern physics, explaining how the sun burns and how electrons move through the tiny circuits in our computers. For decades, physicists have relied on a standard method to calculate how likely a particle is to tunnel through a wall. This method works exceptionally well when the wall is thick and high, making the journey difficult and rare. However, this approach breaks down when the wall is extremely thin, even if it is very tall. In these specific, thin-barrier situations, the old rules fail to capture how the particle moves over time, leaving a gap in our understanding of how quantum systems evolve from a state of being trapped to a state of being free.

A team of physicists has now filled this gap by developing a new way to analyze tunneling through these tall, thin barriers. Rather than treating the barrier as a solid block of matter, they modeled it as a sharp, needle-like spike in the landscape of energy. By treating the strength of this spike as a small, adjustable knob, they were able to track exactly how a particle escapes from a confined space. Their work reveals that the speed at which a particle escapes depends entirely on the symmetry of the space it is trapped in. When the trap is perfectly balanced, with the particle able to reach the other side with equal ease, the chance of finding the particle outside grows rapidly at first, following a curve that accelerates with time. This happens because the particle's energy levels on both sides of the barrier match up perfectly, allowing it to resonate and transfer energy efficiently.

However, the story changes completely if the trap is uneven. In an asymmetric setup, where the space on one side of the barrier is different from the other, the energy levels no longer match. In this case, the particle cannot resonate, and the probability of it escaping grows slowly and steadily, increasing in a straight line with time. The researchers found that this linear growth is a distinct intermediate stage that occurs after the initial burst of activity but before the particle has had time to bounce back and forth enough to create complex patterns. This distinction is crucial because it shows that the shape of the container dictates the rhythm of the escape, a detail that the older, standard methods of calculation simply miss.

The team also explored what happens when a particle is trapped between two such barriers and eventually breaks free to wander off into infinity. In this scenario, the particle behaves like a prisoner rattling against the walls of a cell, with a tiny chance of slipping through the door with every collision. The researchers calculated that the time it takes for the particle to escape follows a predictable pattern based on how often it hits the walls and how thin the barriers are. They were able to write down a complete description of the particle's wave function, showing not just the odds of escape, but the actual shape of the wave as it leaks out and spreads into the open space. This level of detail allows for a clear picture of the transient moments right after the particle begins to move, a phase that is often too fleeting for other methods to catch.

By focusing on these tall, thin barriers, the researchers have created a unified framework that connects three different types of quantum behavior: the smooth transfer of a particle between two identical rooms, the slow leak of a particle from an uneven trap, and the eventual escape of a particle into the vast unknown. Their findings suggest that while the old methods work for thick, difficult barriers, a different, more direct approach is needed for thin ones. This new perspective offers a clearer view of the mechanics of quantum escape, providing a tool that could help engineers design better systems for controlling atoms in ultra-cold environments or understanding how particles leak from microscopic traps. The work confirms that in the quantum realm, the geometry of the trap is just as important as the height of the wall, determining whether a particle slips away in a sudden rush or a slow, steady stream.

Technical Summary: Time-Dependent Tunneling in the Thin-Barrier Limit

Problem Statement
Standard quantum tunneling analysis typically relies on the WKB approximation, which is valid when the tunneling action $S$ is large ( $S \gg 1$ ), corresponding to tall, wide potential barriers. This approach yields time-independent transmission probabilities but obscures the dynamical evolution of the tunneling process and its dependence on the initial state. Conversely, for tall, thin barriers, the action is small ( $S \ll 1$ ), rendering the WKB approximation inapplicable. While previous work has addressed the transmission of plane waves through such barriers using Moshinsky functions, there is a need for a unified perturbative framework to analyze the time-dependent decay of bound states, resonant coherent transfer, and escape into a continuum in this small-action regime.

Methodology
The authors develop a perturbative analysis where the control parameter is $\alpha$ , defined as the inverse of the area under the potential barrier (specifically, the integrated strength of the barrier). The limit of a "tall, thin" barrier is modeled as a delta function potential, $V(x) = \alpha^{-1}\delta(x)$ , where $\alpha \to 0$ represents the opaque limit and the expansion is performed in terms of the barrier's transparency.

The general formalism (Section II) involves:

Solving the Schrödinger equation for $\alpha = 0$ (no tunneling) to establish a basis of symmetric and antisymmetric eigenstates.
Treating the delta function barrier as a perturbation to these eigenstates.
Expanding the initial localized state in the perturbed basis and evolving it in time to derive the time-dependent wavefunction $\psi(t, x)$ and tunneling probability $P_R(t)$ .

This framework is applied to three distinct scenarios:

Symmetric Potentials: Double wells with identical geometry (Sections II & III).
Asymmetric Potentials: Double wells with irrational width ratios, preventing energy level degeneracy (Section IV).
Escape to Continuum: A particle trapped between two delta barriers escaping to infinity (Section V).

Key Results

Symmetric Double Wells (Resonant Case):
For symmetric potentials (e.g., infinite square wells or quadratic potentials) where the left and right wells are identical, the energy levels are degenerate in the absence of the barrier. The tunneling probability $P_R(t)$ for a particle initially localized in the left well grows quadratically with time at early stages:
$P_R(t) \propto \alpha^2 v^2 \left(\frac{t}{\tau}\right)^2$
where $v$ is the particle velocity and $\tau$ is the "rattling time" (time between barrier encounters). This quadratic growth arises from coherent oscillations between nearly degenerate states.
Asymmetric Double Wells (Non-Resonant Case):
When the wells have different widths (specifically an irrational ratio), the energy levels do not match. The resonant enhancement is suppressed, and the tunneling probability grows linearly with time in the intermediate regime (after the initial quadratic transient but before recurrences):
$P_R(t) \propto \alpha^2 v^2 \frac{t}{\tau_L}$
The rate depends on the mismatch between the trapped state's energy and the closest available state in the adjacent well.
Escape to Infinity (Metastable Decay):
For a particle trapped between two delta barriers, the system couples to a continuum. The authors derive the time-dependent wavefunction both inside and outside the trap. The survival probability inside the trap exhibits exponential decay dominated by a metastable pole:
$P_{in}(t) \approx e^{-t/t_{decay}}$
The decay time is given by $t_{decay} = \tau^* / (\alpha^2 v^{*2})$ , where $\tau^*$ is the rattling time and $v^*$ is the characteristic velocity. The wavefunction outside the trap is explicitly constructed, showing the buildup of the transmitted wave.

Significance and Claims
The paper does not claim to introduce new special functions or analytic machinery, as delta-function barriers and time-dependent transients have been studied previously. Instead, the authors claim significance in providing a simple, unified asymptotic organization for several tunneling problems often treated separately.

By utilizing the tall/thin barrier limit as a controlled perturbative expansion, the work demonstrates how:

Resonant coherent transfer (quadratic time dependence) emerges from level splitting in symmetric systems.
Non-resonant leakage (linear time dependence) arises from the absence of level matching in asymmetric systems.
Exponential decay (metastable escape) results from coupling to a continuum.

The authors argue that this approach offers a transparent, elementary framework to compare these distinct regimes (resonant, non-resonant, and continuum escape) within a single formalism, bridging the gap between Green's function pole analysis and Moshinsky function descriptions of quantum transients. The results serve as benchmarks for engineered systems with narrow, tunable barriers, such as ultracold atom systems and mesoscopic structures, where transient transfer estimates are needed without full numerical time evolution.

More like this