Computational Fluid Dynamics (CFD) Modeling of Bubble Dynamics in the Aluminum Smelting Process

Abstract

This paper presents a microscale modeling approach for investigation of bubble dynamics in the aluminum smelting process. The motion of a single bubble has been studied through a computational fluid dynamics (CFD) model facilitated with the volume-of-fluid (VOF) method to capture the bubble shapes. Using a two-dimensional geometry of part of a real cell as the testing bed, the motion of different sized bubbles has been simulated in an air–water system and a CO2–cryolite system. Comparisons between the two systems are conducted through the three periods of bubble motion: bubble sliding under the anode, bubble releasing at the anode edge, and bubble rising in the side channel. It was found that both systems show similar trends in bubble dynamics, such as an increase in the bubble sliding velocity as the bubble size increases and the appearance of a thick head at large bubble sizes. Quantitatively, there are differences between the two systems, evidenced in terms of the detailed bubble dynamics at each period of bubble motion, such as the bubble morphology, the bubble sliding velocity, the bubble layer thickness, and the bubble-induced liquid flow. The detailed microscale modeling provides useful information for the development of a multiscale modeling methodology by building constitutive correlations to support the macro/process scale modeling.