Quantum reflection time and the Goos-Hänchen effect

D. Sokolovski a,b,c,∗, Y. Caudano d · 2025

Read Quantum reflection time and the Goos-Hänchen effect by D. Sokolovski a,b,c,∗, Y. Caudano d (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-17883.txt.

Introduction

Sokolovski and his co-authors present a 2025 physics paper on what it means to time a quantum particle's reflection. When a wave packet bounces off a barrier, it can seem delayed, and a sideways shift known as the Goos-Hänchen effect appears for angled beams. The authors compare tracking a packet's motion with treating the delay as the result of a quantum measurement.

Turning spatial delays into time spans, they argue, can produce misleading negative times and should be avoided. They also show a soft, smooth barrier can be rebuilt as a stack of hard walls. That trick lets them compute the sideways shift for a beam hitting a step or barrier at an angle. The work clarifies how reflection delays should be read in quantum mechanics. It touches debates about tunneling times and wave packet motion.

Catalog record

CREATORD. Sokolovski a,b,c,∗, Y. Caudano d
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE50 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-17883

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Quantum reflection time and the Goos-Hänchen effect D. Sokolovski a,b,c,∗, Y. Caudano d Abstract We explore the analogy between following the motion of a reflected wave packet, and a quantum measurement of the spatial delay imposed on the particle by the scattering potential. It is shown that converting such delays into temporal durations can lead to “negative times” , and best be avoided. It is also demonstrated that a “soft” potential can be replaced by superposition of hard walls. This representation is used for calculating the Goos-Hänchen shift of a Gaussian beam incident on a potential step or a barrier at an oblique angle. I Introduction The Goos-Hänchen (GH) effect, first observed in 1947 for electromagnetic waves GH0 , consists in lateral displacement of a beam reflected off a plane surface. The effect is not unique to electromagnetism, and can occur in a variety of wave phenomena, including quantum scattering, see for example GH2 - GH4 . The displacement is often related with the time delay, associated with the scattering process GH1 , and can be seen as a way of measuring it. However, there is still little consensus about the definition of the quantum mechanical time delay. Much attention has been paid to the “tunnelling time problem” REV1 - REV2 of defining the duration a tunnelling particle is supposed to spend in the barrier region. Recently we argued that the answer is provided by the measurement theory, in a uniquely quantum way DS1 , DS2 , DS3 . According to the Uncertainty Principle FeynL , to know the duration of interest one needs to destroy the interference, and that destroys tunnelling. Conversely, one can preserve tunnelling, but with interference intact the duration of interest must remain indeterminate DS1 , DS2 . The same should be true in quantum reflection, where one would expect similar non-classical behaviour for a particle repulsed by a potential well, or reflected above a potential barrier. In this work, we revisit the quantum Goos-Hänchen effect in order to establish whether determination of the GH shift does amount to measuring the time spent by the particle in the scattering potential. We will answer the question by the negative, and proceed to determine what is being measured, and to what accuracy. The paper is organised as follows. In Section II, we will rewrite the reflected wave packet (WP) in terms of the freely propagating counterpart, and a number of its mirror images. In section III we exploit the analogy with a von Neumann quantum measurement performed on a pre- and post-selected quantum system, and identify the measured quantity. In Sect. IV we show that in the classical limit one does determine the duration spent by the classical trajectory in the scattering potential. In Sect. V we attempt to extend the idea to the fully quantum case, and fail. In Sect. VI obtain the reflection “phase time” and discuss its properties. Sects. VII and VIII offer two simple examples of what has been said so far. In Sect. IX we abandon the attempts to define a meaningful “reflection time” , and replace instead a soft potential barrier by a superposition of virtual hard walls. In Sect. X, we use this representation to reconstruct the reflected beam, and evaluate the GH shift in terms of the spatial shifts experienced by the reflected particle. Section XI contains our conclusions. II Reflection in the momentum and coordinate representations We consider a potential potential V(x) , contained in the region 0<x<d , and a wave packet (WP) state |\Psi(0){\rangle} , {\langle}x|\Psi(0){\rangle}\equiv 0 for x>0 . Figure 1: a) A wave packet with a mean momentum p , incident on a potential barrier V(x) , is partly reflected. Its mean energy is lower than the barrier height, and x_{<} denotes the classical turning point V(x_{<})=E(p) . b) Freely propagating WP and its reflection is a mirror placed at x=x^{\prime}/2 . c) The reflected WP shown in (a) is a superposition of the WPs reflected by a variety of infinite walls. A convenient choice is a Gaussian WP moving towards the barrier (see below). The initial state at t=0 can be expanded in plane waves, (we use \hbar=1 ) \displaystyle\Psi(x,0)=\int_{-\infty}^{\infty}dk{\langle}x|k{\rangle}{\langle}k|\Psi(0){\rangle}, (1) where {\langle}x|k{\rangle}=(2\pi)^{-1/2}\exp(ikx) . Time evolution of the WP is found by expanding it in the complete set of scattering states, incident on the barrier from the left and from the right (see Appendix A). Thus, for t>0 the wave function in the left half-space x<0 is given by \displaystyle\Psi(x,t)=\Psi_{0}(x,t)+\quad\quad\quad\quad\quad\quad\quad\quad\quad\quad\quad\quad\quad\quad (2) \displaystyle\int_{-\infty}^{\infty}\frac{dk}{\sqrt{2\pi}}R(k){\langle}k|\Psi(0){\rangle}\exp[-ikx-iE(k)t],\ \ [x<0], where \displaystyle\Psi_{0}(x,t)=\frac{1}{\sqrt{2\pi}}\int_{-\infty}^{\infty}dk{\langle}k|\Psi(0){\rangle}\exp[ikx-iE(k)t], (3) is the freely evolving initial state, and R(k) is the reflection amplitude for a particle with momentum k , incident on the potential from the left. The function R(k) is known to be analytic in the complex k -plane, with poles in its lower half, and also on the positive imaginary axis, if V(x) supports bound states BZP . Note that Eq.( 2 ) is valid even when if the initial state ( 1 ) contains negative momenta, in which case one uses R(-|k|)=R^{*}(|k|) (see Appendix A). One can transform Eq.( 2 ) to the coordinate representation by inserting the identity \displaystyle R(k)=\int_{-\infty}^{\infty}dx^{\prime}\exp(ikx^{\prime})\xi(x^{\prime}) (4) \displaystyle\xi(x^{\prime})=(2\pi)^{-1}\int_{-\infty}^{\infty}dk^{\prime}\exp(-ik^{\prime}x^{\prime})R(k^{\prime}) (5) into Eq.( 3 ). This yields \displaystyle\Psi(x,t)=\Psi_{0}(x,t)+\int_{-\infty}^{\infty}dx^{\prime}\Psi_{0}(x^{\prime}-x,t)\xi(x^{\prime}). (6) where \Psi_{0}(x^{\prime}-x,t) is the image of the freely propagated WP ( 3 ) in a mirror placed at x=x^{\prime}/2 (see Fig.1b). Each image enters the superposition ( 6 ) with a weight \xi(x^{\prime}) , the Fourier transform of the reflection amplitude with respect to particle’s momentum. For a potential V(x) , such that R(k)\to 0 as |k|\to\infty , \xi(x^{\prime}) is a smooth function of x^{\prime} , and, in the absence of bound states, one has \displaystyle\xi(x^{\prime})\equiv 0\quad\text{for}\quad x^{\prime}<0. (7) The evolution of the wave function in the left half-space can now be visualised as follows. At first, the WP localised far to the left, moves freely towards the scatterer, and the wave function is given by the first term in Eq.( 2 ), \Psi(x,t)\approx\Psi_{0}(x,t) . For a while its mirror images \Psi_{0}(x^{\prime}-x,t) , moving in the opposite direction, remain “off screen” in the right half-space x>0 . As the free WP crosses the origin, and disappears from the screen, its mirror images enter the region x<0 , replacing there \Psi_{0}(x,t) . The scattering is finished when the left half-space contains only the reflected (R) part of the wave function \displaystyle\Psi(x,t){\xrightarrow[t\to\infty]{}}\int_{-\infty}^{\infty}dx^{\prime}\xi(x^{\prime})\Psi_{0}(x^{\prime}-x,t)\equiv\Psi_{R}(x,t).\quad\quad (8) Needless to say, the representation ( 6 ) has no physical significance in the right region x>0 , but is exact for any x<0 . III Analogy with a quantum measurement Let the initial state of a particle of mass m be a Gaussian WP with a mean momentum p=mv>0 , and coordinate width \Delta x , placed a x=x_{0}<0 . Throughout the paper we will neglect the WP’s spreading, so the free state in Eq.( 3 ) propagates without distortion, (see Appendix B), and we have \displaystyle\Psi_{0}(x,t)\approx G_{0}(x-x_{0}-vt)\exp\left[ip(x-x_{0})-iE(p)t\right], (9) where \displaystyle G_{0}(x)=\left(\frac{2}{\pi\Delta x^{2}}\right)^{1/4}\exp\left(-\frac{x^{2}}{\Delta x^{2}}\right). (10) Reflecting \Psi_{0}(x,t) about the point x^{\prime}/2 requires changing x\to x^{\prime}-x , and for the reflected state ( 8 ) we find \displaystyle\Psi_{R}(x,t)\approx\exp[-ip(x+x_{0})-iE(p)t]\times\quad\quad\quad\quad\quad (11) \displaystyle\int_{-\infty}^{\infty}dx^{\prime}G_{0}(x^{\prime}-X(t))\exp(ipx^{\prime})\xi(x^{\prime}),\quad[x<0], where we introduced a shorthand \displaystyle X(t)\equiv x+x_{0}+vt. (12) Similarity with a quantum measurement made on a pre- and post-selected quantum system

Read the complete text