Copper Corrosion Mechanisms by Simulation and Experiment Using Small Test Structures

Hasan Sadat Nabi1, M. Mayr2
1Infineon, Regensburg
2TU Wien
Publié en 2023

Product lifetime prediction is a crucial task in microelectronics design and fabrication. Understanding of the underlying processes causing material degradation and product failure is essential for establishing accurate lifetime models. This work aims to investigate and model the degradation of organic coatings (OCs), which are commonly employed to protect microelectronic components from external influences (e.g. bias, humidity), as well as the electrochemical migration (ECM) processes in/on coated structures, such as copper (Cu) metal conductors. ECM processes depend very much on the properties of the OC and its degradation behavior over time in a humid environment [1]. These processes end up in dendrite growth, causing short circuits. Two experimental setups are used for variation of electrolytes and application of bias to ‘scan’ migration and corrosion mechanisms. Electrochemical impedance spectra and current-versus-time curves are recorded during the experiments (Fig. 1), along with optical images. The COMSOL Multiphysics® software is used for simulations of diffusion and migration by solving Nernst-Planck equations as well as electrochemical and homogeneous reactions (tertiary current distribution) in an electrolyte. Experimentally, the influence of different electrolytes and their concentrations, as well as applied biases on the corrosion behavior are determined along with the conditions in which dendrites form. Overall agreement between experiment and simulation is good, but still has to be optimized. Deviations are mainly attributed to a missing implementation of actual dendrite growth.

Téléchargement