An Extendible Multiphysics Model of a Single Water Electrolysis Cell

Abstract

Listen

Translate article

The important role that renewable hydrogen has in the energy transition is undeniable given the capability of decarbonizing sectors difficult to electrify acting as the link between electrical energy from renewable sources and molecules for future energy systems. Therefore, improvement of the efficiency, cost and durability of electrolyzer systems is key. The discovery of new materials is one of the research areas in which a lot of effort is focused given that finding more electrochemically active materials allows to reduce overvoltages improving the electrical performance of the electrolyzers. The requirement for these materials is also that they are durable under operating conditions and cheap to produce in order to achieve improvements in all relevant aspects of the technology.In water electrolysis systems, kinetics, heat management, gas quality, among others, are relevant aspects to understand in order to achieve performance gains. In this line, computational modeling plays an important role in analyzing and understanding the performance of such systems, as well as the effect that different variables have on it. Given the nature of the water electrolysis process, a multiphysics approach is necessary to account for electrochemical, mass transfer, thermal and fluidic processes, as well as their interactions.Computational models and digital twins are gaining great momentum in electrolysis systems applications, but its availability is limited as its development requires of deep physical knowledge of the different phenomena and their interactions, or large amounts of data. When data is available data-driven models are very useful but the results are not easy to extrapolate when a change in design or materials happens. Therefore, physics-based modelling is a useful tool, as characterization at small scale and short-term is the first step in the discovery of new materials, which allows to obtain physical parameters that act as input for the computational model. This way it is clear that the advantage of developing a physical model is that the model becomes generic, which means that by changing the corresponding design parameters, the performance of a different design can be analyzed.The current project aims to develop an extendible physics-based model of a single water electrolysis cell. This development will allow to modify different parameters or operating conditions of the cell design and analyze the effect of those changes in the performance. Additionally, the extendible nature allows for its use as single element for analyses in 1D if desired.The model, developed in OpenModelica, is capable of reproducing the polarization curve of an electrolysis cell, the temperature change between inlet and outlet, the gas crossover and the pressure loss.