Comparison of different performance recovery procedures for polymer electrolyte membrane fuel cells

Abstract

Listen

Translate article

• Quantitative analysis and comparison of three different recovery procedures. • All procedures yield high recovery in mass transfer resistance. • Short H 2 soak insufficient to recover increase of charge transfer resistance. • Careful dry gas purge crucial to avoid increase of ohmic resistance. • Quantification of performance recovery needs to consider operation history. To both extend system lifetime and enable reliable performance benchmarking, recovery procedures are of great importance to distinguish and mitigate the reversible and irreversible performance degradation of polymer electrolyte membrane fuel cells. In this work, three common recovery procedures available in the literature (JRC-based protocol, DOE-based protocol, and overnight rest) being part of a load cycling durability test are characterized by polarization curves and electrochemical impedance spectroscopy. Such direct comparison using same conditions and material has not been published so far. To compare the relative recovery of the three recovery procedures the recovery related to the last operation period, to the beginning of the whole test, and the non-recovered performance loss within each operation period are assessed. Moreover, the mechanisms leading to the various recovery effects are analyzed. Generally, with the contribution of gas purging to water removal, all three recovery procedures reduce greatly mass transfer resistance and recover most of the performance loss in the high current density range. At lower current density, the three procedures differ substantially. In the case of JRC-based protocol, the kinetic losses are recovered by the reduction of Pt oxides and structure change of ionomer by reducing the cathode potential and fuel cell temperature, respectively. The overnight rest results in similar recovery of the performance loss. The DOE-based protocol leads to relatively low recovery of losses in the kinetic region of the polarization curves. Additionally, the effect of the cell operating history is considered.