Shot noise in resonant tunneling: Role of inelastic scattering

TL;DR

This paper investigates how inelastic scattering affects shot noise and the Fano factor in resonant tunneling through a double-barrier structure, deriving a general formula and analyzing phonon-induced dephasing effects.

Contribution

It introduces a Landauer-Büttiker-type formula for current noise incorporating inelastic scattering amplitudes and calculates exact scattering amplitudes for phonon-induced dephasing.

Findings

Fano factor is insensitive to inelastic processes when multiple levels are within the voltage window.

At low voltages, the Fano factor depends strongly on inelastic scattering, invalidating phenomenological models.

Resonant level shapes vary with phonon dimensionality, affecting shot noise characteristics.

Abstract

We study the influence of inelastic processes on shot noise and the Fano factor for a one-dimensional double-barrier structure, where resonant tunneling takes place between two terminals. Most studies to date have found, by means of various approximate or phenomenological methods, that shot noise is insensitive to dephasing caused by inelastic scattering. In this paper, we explore the status of this statement by deriving a general Landaur-B\"uttiker-type formula that expresses the current noise and Fano factor in a one-dimensional conductor through inelastic scattering amplitudes. For a double-barrier structure, exact scattering amplitudes are calculated in the presence of a time-dependent potential. As an example of dephasing potential, we consider the one induced by equilibrium phonons. We calculate transmission coefficients of a double-barrier structure for these two types of phonon-induced dephasing. In the case of diffusive phase relaxation valid for one dimension phonons, the resonant level has a Lorentzian shape. For phonons whith high dimensions logarithmic dephasing realized which leads to an unusual shape of the size-quantized level characterized by the two energy scales. We further calculate the Fano factor for these types of dephasing, using exact expressions for inelastic transmission and reflection amplitudes. It turned out that when an integer number of levels fall into the energy window of width eV, where V is the voltage applied to the structure, the Fano factor is really insensitive to inelastic processes inside the structure and coincides with the prediction of phenomenological models with an accuracy of small corrections depending on these processes. On the contrary, at low voltages, when the eV window is smaller than the level width, this dependence is particularly pronounced and the phenomenological formula does not work.