Flow investigation of complex suspension electrodes for battery applications using ultrasound imaging velocimetry

Abstract

Flow batteries using suspension electrodes are currently investigated due to their potentially high energy densities and individual scalability of capacity and power density. In contrast to batteries with statically contacted electrodes or conventional redox flow batteries, their performance depend significantly on the local flow conditions of the electrode suspension in the electrochemical cell. Hence, it is crucial to understand and model the complex flow characteristics of such solid-liquid suspensions.Detailed ultrasound flow measurements are combined with multidimensional modeling of the bulk rheology and wall interactions for opaque, thixotropic multiphase fluids in a real-world flow setup. The present flow investigations are conducted for zinc-air type of batteries using a gelled alkaline electrolyte with suspended microscopic zinc particles, however the utilized methods could be adapted to other complex suspension flows of interest. Due to the opacity of the suspension, an ultrasound flow mapping technique, the ultrasound imaging velocimetry (UIV), was applied. In a reference measurement using the model fluid glycerol, the UIV is validated by comparing the measured flow field to optical particle image velocimetry (PIV) measurements and a numerical simulation.In contrast to a Newtonian fluid, the gelled suspension with 8 vol.% particle loading exhibits significant time-dependent non-Newtonian flow characteristics with slip at the wall. While initially strongly sheared zones retain a comparatively low viscosity and allow for high velocity gradients, in areas with low initial shearing (e.g. in corners) the fluid structure is able to rebuild, resulting in a solid-like behavior. An in-depth understanding of the development and the extent of such zero-flow zones can be critical for reliable and efficient designs of the relevant fluidic structures. The measurement data of the zinc-electrolyte suspension flow were used as reference to estimate parameters for a computational fluid dynamics model considering the time-dependent, non-Newtonian rheology of the bulk fluid and the apparent slip at the walls. The simulation model is able to predict the main characteristics of the measured two-dimensional flow fields. Despite some uncertainties, the methods prove to be effective for investigating complex suspension flows and they can be applied to improve and optimize the corresponding fluidic structures, either in flow batteries with flowing suspension electrodes or other industrial applications, including the processing of polymers, minerals, cosmetics, food products, pharmaceuticals and ceramics.