PhD Thesis Defence Presentations - Panagiotis Giotakos

In today’s modern era, the need for decarbonisation of energy production is imperative for the sustainable development and prosperity of human society. The development of phosphoric acid doped polybenzimidazole (PBI) electrolyte membranes in 1995 allowed the operation of Polymer Electrolyte Membrane (PEM) fuel cells (FC) at temperatures higher than 100^oC. This led to the development of High Temperature Polymer Electrolyte Membrane Fuel Cells (HT-PEMFC), one of the most potent electrochemical devices for efficient clean power production. Since then, there has been extensive work done on developing and optimizing HT-PEMFCs for improved performance, extending the lifetime of operation, reducing manufacturing costs and most importantly increasing the understanding of the electrochemical processes occurring during operation.

The aim of this dissertation is to provide a deeper understanding of the electrochemical processes inside the catalytic layers of the two electrodes, focusing especially on the cathodic electrode, where Oxygen Reduction Reaction (ORR) takes place. ORR mechanism has been under the microscope for more than five decades, since it is solely responsible for introducing the largest potential losses, which in order to mitigate require high amounts of expensive Pt catalyst loadings.

To this end, mathematical modeling (both numerical and analytical) is successfully coupled with two electrochemical characterization methods; the steady state polarization (IV) curve characterization method and the Electrochemical Impedance Spectroscopy (EIS) characterization method. Various parameters that affect the performance of  HT-PEMFCs are studied by applying IV-EIS, however the main contribution of this thesis comes from unraveling the sluggish kinetics and energetics of the so far elusive ORR. 

We performed meticulous/detailed experimental steady state (IV) and EIS measurements in the activation region of operation by carefully controling the experimental operating conditions, in order to avoid mass transport limitations and excessive water production, which would severely affect the structure of the Cathodic catalytic layer (CCL). 

By experimentally observing that the limiting current density of the FC was controlled by kinetics (instead of mass transport), we employed the simplest ORR mechanism (the Dissociative Pathway) using microkinetic modeling based on the Transition State Theory. For the representation of the CCL a macro-homogeneous Transmission Line Model (TLM) was used and both the IV and EIS were derived analytically, (with only two somewhat arbitrary assumed parameters i.e., the symmetry factor β = 0.5 and Langmuir isotherm kinetics). We were able not only to achieve excellent fitting results (using our in house developed Monte Carlo Least Squares Fitting Algorithm) with the experimental data (IV & EIS), but to successfully extract ORR kinetics (such as kinetic constants, double layer capacitance, ionic conductivity etc.) and calculate the energetics of the elementary steps, i.e.  both the activation energies (kinetics) and the reaction steps’ free energies (thermodynamics), bridging the gap between Density Functional Theory equilibrium analysis and electrochemical kinetics dynamic analysis.

The excellent fitting results revealed that: i) all charge transfer reaction steps appear under the same high frequency arc,  directly related to the Cdl of the electrochemical interface, while the low frequency arc originates from the ORR kinetic inertia, ii) both EIS and polarization resistance are dominated by the intrinsic ORR kinetic inertia due to the competitive nature of the elementary reaction steps on the coverages of the adsorbed species (O_ad and OH_ad), iii) with the help of Degree of Rate Control (DRC) analysis we identified O2_(g) dissociative adsorption as the main limiting step responsible for the introduction of high ORR kinetic and thermodynamic overpotentials, iv) the independent calculation of H2_(g) crossover current density from the anode side was in excellent agreement with our model predictions and proved that it is the one which solely responsible for the deviation of the experimental open circuit potential from its thermodynamic Nernst value. 

Last but not least, the contributions of this thesis not only enrich the scarce literature of successfully extracting kinetic and energetic information of an actual HT-PEMFC but also provide a guide, a methodology for screening and optimizing electrocatalysts for multistep electrochemical reactions.