Multiphysics modelling and experimental validation of water crossover phenomena in Vanadium Redox Flow Batteries

Flow batteries (FBs) are a type of electrochemical energy storage system in which electrolyte solutions, stored in external tanks, are circulated through an electrochemical cell during operation. This distinctive architecture decouples energy storage capacity—governed by the volume of the tanks—from power output, which depends on the number and size of cells in the stack. Among various flow battery technologies, Vanadium Redox Flow Batteries (VRFBs) have emerged over recent decades as the most studied and developed, positioning themselves as a leading candidate for large-scale, long-duration energy storage (LDES), particularly in support of renewable energy integration. Despite their advantages, the long-term performance and reliability of VRFBs are hindered by several parasitic phenomena that reduce efficiency and require careful system management. Among these, crossover phenomena—specifically the undesired transport of species through the ion-exchange membrane—play a critical role. While ionic crossover of vanadium species has been extensively studied, water crossover remains a relatively underexplored yet impactful mechanism. The transport of water between half-cells causes progressive volume imbalances in the electrolyte tanks, which can compromise cell performance, decrease energy efficiency, and necessitate periodic corrective actions. This thesis addresses this gap by incorporating water crossover dynamics into an existing multiphysics VRFB model developed in the COMSOL Multiphysics® simulation environment. The enhanced model captures the evolution of electrolyte volumes in both tanks during charge-discharge cycles, coupling electrochemical kinetics, mass transport, and fluid dynamics to provide a comprehensive and physically consistent description of internal system behavior under realistic operating conditions. Validation was conducted through experimental tests at the Electrochemical Energy Storage and Conversion Laboratory (EESCoLab), where volume variations were monitored during controlled cycling. These data were used to calibrate key parameters and verify the model’s predictive accuracy. Beyond the single-cell 2D configuration, the model was extended to simulate a 2D stack and a full 3D stack, enabling more realistic analysis of practical VRFB systems. By integrating water crossover effects and scaling the simulation to more complex geometries, this work enhances our understanding of operational imbalances in VRFBs and provides a valuable tool for designing more robust, efficient, and self-balancing flow battery architectures.