arXiv:2610.32121
Material criteria for angular-momentum transfer from twisted photons to photoelectrons
A. A. Shchepkin, D. V. Karlovets
Abstract
We study the material requirements for transferring the total-angular-momentum (TAM) projection of a twisted photon to the orbital angular momentum (OAM) of a photoelectron emitted from an $n$-type semiconductor surface. Within a single-valley effective-mass model of surface photoemission, we calculate OAM-resolved spectra for GaAs, InSb, $\mathrm{InAs}_{0.34}\mathrm{Sb}_{0.66}$, and $\mathrm{Hg}_{0.76}\mathrm{Cd}_{0.24}\mathrm{Te}$ irradiated by twisted photons with Laguerre--Gaussian--like momentum profile. The OAM transfer is characterized by the mean electron OAM, its root-mean-squared width, and the fraction of electrons with OAM-projection equal to the photon TAM-projection. Efficient transfer requires the transverse momentum of the conduction electron to be much smaller than the transverse momentum width of the photon, which favors low temperatures, small conduction-band effective masses, and shallow donor levels. GaAs is less favorable within the present model because of its larger effective mass and deeper donor levels. We also discuss the demanding photon-wave packet parameters, possible extensions to Dirac and Weyl semimetals, and post-emission OAM selection. Although the generation of vortex electrons with a vanishing OAM uncertainty via photoemission seems challenging, we argue that it is still possible to generate single-electron vortex states with a finite OAM dispersion.
AI Summary
Can twisted-light photoemission prepare electrons with a definite orbital-angular-momentum (OAM) projection, and which semiconductor properties favor this transfer? The paper applies an existing surface-photoemission theory to weakly doped $n$-type GaAs, InSb, $\mathrm{InAs}_{0.34}\mathrm{Sb}_{0.66}$, and $\mathrm{Hg}_{0.76}\mathrm{Cd}_{0.24}\mathrm{Te}$. Numerical calculations use a single-valley, parabolic effective-mass model, thermal conduction-electron populations, a flat interface, and Laguerre–Gaussian-like photon wave packets. OAM spectra are evaluated at selected outgoing momenta near the detection-probability maximum. Exact transfer $l_e=m_\gamma$ occurs for zero initial transverse electron momentum; selective transfer requires that momentum to be small relative to the outgoing transverse momentum and photon transverse width. Low temperature, small effective mass, and shallow donors favor this hierarchy. With photon widths $\sigma_\perp=1$ eV and $\sigma_z=2$ eV, the desired-OAM fraction at $0.5$ K is $87$–$91\%$ for $m_\gamma=1$ across the three narrow-gap materials. For $m_\gamma=64$, their mean OAM remains $63.0$–$63.4$, but only $16$–$17\%$ occupies the desired channel. At $4$ K, that fraction becomes $8$–$15\%$, with rms widths near $3$. GaAs is less selective within this model. The normalized spectra do not establish absolute yield or intermode coherence. The model excludes explicit spin dynamics, scattering, surface disorder, and activated negative-electron-affinity cathodes; nonparabolicity also limits applicability. The required broadband, tightly focused UV vortex packets are experimentally demanding. Downstream OAM sorting and postselection are discussed, not demonstrated; Dirac/Weyl materials require a separate multiband treatment.
Error found: The reported desired-OAM fractions are conditional on selected final electron momenta and do not establish the fraction in the desired OAM channel across the emitted beam or a finite experimental acceptance. The adopted InAs0.34Sb0.66 mass, hydrogenic donor energy, and donor concentration give n_d^(1/3)a_d* ≈ 0.39, inconsistent with the paper's stated isolated-donor criterion of approximately 0.25 or less.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 15 pages, 3 figures