SOFC Degradation Quantification Using Image Analysis

摘要

SOFC cell degradation is, amongst others, related to changes in the morphology and chemical composition of cell layers, to the deposition of impurities and chemical species in layers, or the reaction of compounds with each other or emissions from other SOFC stack components. Up to today, the dependency of the actual loss of power (or voltage) on specific changes in morphology or composition are only qualitatively known. The existing models use empiric or statistic evaluation of responses of fuel cells to operating conditions in order to fit models of degradation. Obviously, there must be a relationship between the changing structures and compositions of the layers in an SOFC cell with the corresponding degradation mechanisms. Nevertheless, it has proven difficult to quantify this relationship and turn the achieved insight into models predicting degradation. The main problem is the way of turning microstructural data, for instance from microscopy images (SEM, BSE, EDX etc.), into quantified information that can directly be correlated with measured effects. Within the project Image-SOFC, the Institute of High Temperature Electrochemistry (Ekaterinburg), the Institute of Mathematics and Mechanics (Ekaterinburg), and Forschungszentrum Julich have conducted a project developing image analysis methods that are capable of reliably identifying not only pores but also different phases in SOFC cell layers. The method delivers quantitative figures on the morphology and phase distribution that allows direct insertion in equations describing temporal evolution of cell properties. Examples of results that will be presented are the detection of chromium poisoned cathode layers and changes in anode nickel cermet morphology due to nickel agglomeration.