A Computationally-Efficient, Zero-Dimensional Stack Model for Simulating Redox Flow Battery Performance

Redox flow batteries (RFBs) are an electrochemical energy storage platform with potential to support grid-scale decarbonization and resiliency efforts. The RFB architecture allows for decoupled energy and power, long lifetimes with simplified maintenance, and a range of potential chemistries. Despite their promising characteristics, present embodiments are too expensive for widespread adoption. 1 , 2 Significant research efforts have focused on advancing new redox couples and constituent materials (e.g., electrodes, flow fields, membranes), but these contributions primarily focus on individual lab-scale cells, due, at least in part, to the time, material, and expertise required to investigate performance and durability in larger device formats (e.g., multi-cell stacks). This results in knowledge gaps within the field—specifically, electrochemical reactor scaling relationships are not yet well-understood and the relative importance of many tunable molecular and material properties at the stack level remains unclear. To aid with RFB stack design and understanding, computational models of varying dimensionality and complexity have been developed to simulate charge-discharge cycling. 3 Zero- and one-dimensional models, which make simplifying assumptions about underlying electrochemistry and fluid dynamics, are particularly useful for capturing general trends in cell performance at a much lower computational cost than comprehensive multi-dimensional frameworks. However, previous models still employ numerical methods to solve complex systems of differential equations, which slows simulation time and impedes broader inquiry (e.g., durational cycling, parametric property sweeps). 4 ,5 Further, existing models are typically designed to interrogate specific chemistries and, as such, cannot easily incorporate the array of properties and operating conditions possible for different redox couples and constituent materials. In this presentation, we will discuss the formulation and application of a computationally-lightweight zero-dimensional RFB stack model suitable for linking stack-level electrochemical and fluid dynamic performance characteristics to molecular- and cell-level property sets. By constructing analytical expressions for mass balances and cell voltages under galvanostatic cycling conditions, 6 this framework facilitates connections between system design, component material properties, operating conditions, and cycling performance at the stack level with li