Predicting the Formation of Electrodeposition Patterns Caused By Interfacial Instabilities

Abstract

Electrodeposition is a common process via which metal ions are reduced at a cathode as a thin film. However, the deposited film on the cathode often exhibits a unique interfacial pattern that is dependent on the applied voltage, the concentration, and the distance between electrodes. Although these morphological instabilities have been observed in the past, it still remains difficult to predict the interfacial pattern formed during electrodeposition. These morphological patterns arise in an electrolytic cell because the reduction reaction runs faster than oxidation. Therefore small crests or troughs on the surface of the electrode act as a disturbance that alters the rate and uniformity of metal deposition at these sites, initiating the growth of an interfacial pattern. We first develop a model that describes the electrodeposition instabilities that lead to these interfacial patterns. Transport and kinetic reactions described by Butler-Volmer equations are used to calculate and predict the growth characteristics of the film. It is found that the patterns of the film are periodic and harmonic waves composed of different wavelengths. While each wavelength corresponds to a specific growth rate, many wavelengths have similar growth rates. Changes to surface tension and ionic diffusion can yield narrower distributions of the wavelengths observed. A simple Cu-Cu electrodeposition system is then investigated to observe the instabilities. The pattern on the cathode is observed as a function of deposition time in the presence of random disturbances. By image analysis, the wavelength distribution of the interfacial pattern is detected, which is consistent with the developed model. Due to its inherent instability, the harmonic waves of the pattern are not identical every time under the same experiment condition. Nevertheless, the wavelengths observed during deposition are still located within the predicted range. The results demonstrate that the wavelengths of the pattern are predictable but difficult to control because of inherent disturbances on the surface. Further, the growth of the patterns creates localized flow convection that leads to the formation of dendritic patterns. Now that the model can predict these patterns, the goal will be to find experimental conditions that can either minimize pattern formation or generate a single wavelength.