A 2D Modeling Study on the Heat Generation within Carbon Dioxide Electrolysis Systems for Different Geometries

Abstract

An energy transition is needed in order to combat climate change. With the rise of intermittent renewable energy, a need for energy storage is also inevitable. Carbon dioxide electrolysis is a potential solution as CO₂ emissions can be recycled and subsequently converted for energy storage. However, the technology is rather new and research has yet to be conducted in this field. An important aspect is the temperature within an electrochemical cell, especially when scaling up. An increase in temperature can benefit the performance of the cell, but it also has downsides. Hot-spot formation with non-uniform reaction kinetics and thermal sensible components can have a great influence on the life-time of the cell.



For that reason, a modeling study on the heat generation within carbon dioxide electrolysis systems is done. Different volumetric gas flow rates of carbon dioxide have been considered for two geometries: the membrane electrode assembly (MEA) and the gas diffusion electrode (GDE). The model considers three separate models: a mass model, electrochemical model and thermal model, and operates at a fixed current. The finite difference method is applied using Python 3.0 to solve the relevant conservation equations. Furthermore, the model includes different material layers, where the materials and dimensions are based on recently done experiments.



The model showed that irreversible losses caused by the activation overpotentials are the biggest contributor to the total heat generation. Reversible heat also contributes to the heat generation, where heat is required in the anode and heat is generated in the cathode. Furthermore, within the cathodic catalyst layer most heat is generated. Joule heating caused by ohmic losses has proven to have negligibly impact on the total heat generation. As a result, the hot-spot is located within the cathodic catalyst layer for both geometries. Due to the additional electrolyte in the GDE, the hot-spot does not reach the membrane, in contrast to the MEA. Besides, different results in the y-direction are observed for the volumetric flow rates. For both geometries, the hot-spot is located at the inlet for 10 ml/min and in the middle for 100 ml/min. From the analysis, the GDE is more favorable as less heat is expected and the hot-spot does not reach the membrane. The sensitivity analysis showed that the thermal conductivity is of great importance.