Upscaling solid oxide electrolysis cell CFD simulations for hydrogen production

The role of hydrogen production in the energy transition is of fundamental importance, and Solid Oxide Electrolysers (SOE) are expected to become one of the most efficient hydrogen production technologies. Upscaling Solid Oxide Electrolysis Cells (SOECs) while maintaining overall cell performance and reducing degradation still requires significant experimental and capital resources. This study investigates the effects of grid refinement in Computational Fluid Dynamics (CFD) SOEC simulations and the possibilities of accelerating convergence times. The electrochemical reactions, species transport, fluid dynamics, electron transfer, and heat transfer are modelled through the commercially available CFD software FLUENT. The effects of optimisation measures and scale-up are evaluated through the resulting multidimensional temperature, current density, species, and pressure profiles. The analysis of a 25 cm2 active area design reveals the effect of grid refinement on predicting SOEC performance. The results of a 100 cm2 active area industrial-sized SOEC model prove the accuracy of thermoneutral state simulations with low grid densities of 18,404 cells per square centimetre of active area and reveal challenges in technology scale-up. In addition, the comparison of numerical relaxation methods shows that the pseudo-transient relaxation scheme drastically improves the convergence in CFD SOEC simulations, thus enabling a higher computing performance and therefore lower SOEC development times.Highlights Successful formulation of convergence enhancing CFD solver settingIdentification of efficient meshing strategies at thermoneutral stateDirect application to small scale and industrial scale SOEC