The performance of Solid Oxide Fuel Cell (SOFC) electrodes is determined both by their
porous microstructure and the intrinsic properties of their component materials. This thesis details the development of two characterisation tools for analysis of mass and charge transport processes in SOFC electrode materials.
Firstly, a new approach to isotopic exchange is described, which allows the oxygen self-
diffusion (D^∗) and effective surface exchange (k^∗) to be measured in ambient atmospheres.
This is significant as many similar studies in the literature are limited to investigations in
pure, dry oxygen, or other proxy environments, which are not representative of realistic SOFC
operating conditions. A finite difference simulation was created to generate profiles that were used to extract material parameters from this “back-exchange” data. The technique was then validated by comparison of the results from two experiments in pure, dry oxygen (both single step and back-exchange), which demonstrated good agreement with values of D ∗ and k ∗ in the literature, for the common SOFC cathode material La_0.6Sr_0.4Co_0.2Fe_0.8O_3−δ (LSCF6428).
A third experiment found the surface exchange coefficient to increase by a factor of 5 when
exchanged under ambient conditions compared with pure, dry oxygen.
Secondly, following an introduction to the tortuosity factor, X-ray tomography was used to
3D image micro-tubular (MT) samples ( c. 1 mm diameter) at three key length-scales. A zirconia
based MT-Solid Oxide Electrolyser Cell (SOEC) was imaged at the whole cell level, both before
and after 300 hours operation at 750◦ C . Current collector contact was found to be poor even before operation, but afterwards the paste was seen to agglomerate into metallic silver and no longer span the gap to the current collector wire, which further degraded contact. A ceria based MT-SOFC with a hierarchical microstructure was then imaged at both the micro- and nanoscale. The geometry data was used to determine that the tortuosity factor of this radial system was significantly higher when measured at either length-scale in isolation, than when considered together in a multi length-scale model.