The Effect of Interdigitated Channel and Land Dimensions on Flow Cell Performance

Abstract

We demonstrate the effect of varying the channel and land width dimensions of an interdigitated flow field experimentally and through computational modeling. Measured polarization curves (overpotential versus current density) are reported for a symmetric cell with a ferrocyanide-ferricyanide electrolyte in aqueous potassium chloride. A two-dimensional, coupled fluid dynamic and electrochemical model is developed to interpret polarization results from the symmetric cell. This model suggests that stagnant fluid zones above the center lines of the interdigitated channels negatively impact cell performance. A three-dimensional computational fluid dynamic model is used to calculate the pressure drop and power losses associated with flowing fluid through these flow fields. The voltage efficiency of the cell, corrected for pumping power losses, is evaluated and reported as a function of channel and land width to identify high-efficiency flow fields. The implications for the engineering of large-area flow fields are discussed.

Electrolyte Flow Field Variation: A Cell for Testing and Optimization of Membrane Electrode Assembly for Vanadium Redox Flow Batteries

“…This contradicts with previously reported data, stating that interdigitated flow field is superior to serpentine. It was shown by Gerhardt that parameters of channel length and land width have a significant impact on the performance of the cell. Therefore, it is possible that parameters of tested configuration need further optimization.…”

Integrated Numerical and Experimental Data-Driven Parameter Identification of Electrode Properties in All-Vanadium Redox Flow Batteries

“…Similarly, the A V,− quantity was directly input instead of employing the long fiber approximation which has been previously used in simulations [33,48,49]. Indeed, there are multiple methods with which to approximate surface area [105]; however we chose to keep the A V,− quantities unconstrained by equations to promote electrode property exploration.…”

“…To aid in the visualization of the data, a dimensionless quantity was introduced to define a unique variable that is a function of input parameters. Often, this non-dimensionalization can aid in combining multiple simulation inputs into a single quantity [33,48,49]. The dimensionless quantity is defined as follows:…”

“…These simulations typically require measured, fitted, or referenced chemical property data in addition to specified governing equations and geometric dimensions for establishing the computational domain. Three-dimensional (3D) RFB models have been developed and compared to experimental results [14,[30][31][32][33][34][35][36]. However, given the various feature lengths in a RFB, 3D model are computationally expensive, which has, in turn, motivated the development of two-dimensional (2D) RFB models to more quickly compare simulation results with lab-scale cells [37][38][39][40][41][42][43][44][45][46].…”

Data-Driven Electrode Parameter Identification for Vanadium Redox Flow Batteries Through Experimental and Numerical Methods

“…Similarly, the A V,− quantity was directly input instead of employing the long fiber approximation which has been previously used in simulations [33,48,49]. Indeed, there are multiple methods with which to approximate surface area [105]; however we chose to keep the A V,− quantities unconstrained by equations to promote electrode property exploration.…”