Time delay as the origin of oscillations in anodic Si electrodissolution

TL;DR

This paper presents a mathematical model explaining oscillations in the anodic dissolution of silicon, highlighting the role of time delay and defect dynamics, and successfully reproduces experimental behaviors.

Contribution

The study introduces a physicochemical-based mathematical model incorporating time delay, providing new insights into silicon electrochemical oscillations and potentially applicable to other metal or semiconductor systems.

Findings

Numerical simulations match experimental oscillation patterns.

Time delay is essential for oscillation stability.

Model aligns with the defect model for passive film growth.

Abstract

Silicon is the most important semiconductor electrode with applications in photoelectrochemistry and sensor technology. Yet its electrochemistry exhibits many poorly understood phenomena, including oscillations during the anodic dissolution of silicon electrodes. In this article, we present a mathematical model based on physicochemical steps that captures these oscillations and enables a thorough understanding of the underlying mechanism. The model describes the formation and dissolution of an oxide layer, and determines the oxide composition and the electrostatic potential in the direction perpendicular to the electrode. Oscillations occur if the following conditions are fulfilled: 1. The etching speed increases with defects in the oxide layer 2. The number of defects decreases with increasing electric field strength at the Si-oxide interface. 3. There is a sufficient time delay between the production and the etching of the oxide. Numerical simulations reproduce experimental results well, including the dependence of the oscillation amplitude and period on the potential, as well as the hysteresis behavior observed in cyclic voltammetry. Based on these results, we derive a simplified time-delay differential equation model. Using linear stability analysis, we confirm the essential role of the time delay for the oscillatory oxide-layer dynamics. The basic steps of the model are general and in line with the point defect model for growth and dissolution of passive films on metal electrodes. Therefore, it is likely applicable to a variety of oscillating anodic metal or semiconductor dissolution reactions.