Tubular High Temperature PEM Fuel Cell with Novel 3D-Printed Metallic Porous Anode Gas Diffusion Layer: Fabrication, Model Development and Validation

The high temperature proton exchange membrane fuel cell (HT-PEM-FC) is considered a promising candidate for energy generation in stationary and non-stationary applications mostly due to its middle working temperature, compactness and high power density. The plate-frame design is the most widely used cell configuration, in which a significant portion of the manufacturing cost falls on components such as bipolar plates, current collectors and gas diffusion layers. To overcome this problem, researchers have introduced 3D-printing technology into the field for production of bipolar plates and gas diffusion layers to some small extent. However, further cost reduction is expected if alternative fuel cell configurations are investigated. For instance, the well-known tubular design commonly used in the solid oxide fuel cells, in which case the bipolar plates can be omitted.

This paper presents a fabrication method of a single concentric multilayer tubular HT-PEM-FC using a novel 3D-printed metallic porous anode GDL produced via selective laser melting (SLM) method. Furthermore, a two-dimensional axisymmetric non-isothermal steady state model is developed. This model considers the major transport phenomena as well as the fuel cell loss mechanisms, namely activation losses, ohmic losses and mass transport losses, while mixed potential is neglected. The membrane electrode assembly (MEA) was modeled analytically using a thin-layer approach and the Butler-Volmer equation to describe both electrochemical reactions. Additionally, the model was validated versus experimental data and in addition was compared with an available detailed model. The fuel cell was electrochemically characterized using a commercial test station by means of polarization curves with hydrogen as fuel and air as oxidizer.

The developed model shows good agreement with both experimental data and simulation results from the detailed model, demonstrating that it is applicable to predict the complex fuel cell internal processes and fuel cell performance of the fabricated tubular HT-PEM-FC. Differences were observed in some cases at the open circuit voltage (OCV) and lower density current region, which indicates that mixed potential would be required for a better adjustment, especially in degraded fuel cells. Further evaluation of process and operation parameters can be conducted using the developed model without the need of additional experiments. This will lead to an important reduction in cost and time of manufacturing plus a possible optimization of the system. Moreover, the experimental results demonstrated the suitability of SLM as additive manufacturing process for gas diffusion layer construction and opened up the possibility to be creative in the pursuit of unconventional designs in a cost effective manner.