Physicochemical properties evolution of cathodic catalyst layer and cell performance improvement during PEMFC start-up with mitigation strategies

International audience; Proton Exchange Membrane Fuel Cell is a promising technology to power electric vehicle. To mass-market, one of the technological limitations to overcome concerns improvements in durability[1]. Degradation mechanisms during real-life condition in vehicular operation have been reviewed and authors revealed start-up/shut-down (SU/SD) as critical phases [2]. Indeed, the well-identified reverse current decay mechanism occurs during the H2|Air coexistence in anode side and it is recognized to accelerate Platinum catalyst and carbon support degradations within Cathode Catalyst Layer (CCL)[3,4]. Frequent SU/SD results to a progressive evolution of CCL physiochemical properties, and ultimately, cell performance decay. In this work, a novel Accelerated Stress Test (AST) was designed to mimic realistic potential profile at the cathode side during the SU-phase[5]. By comparing to the well-known triangular profile recommended by SU DoE[6], our profile adds a high potential plateau at 1.6 V during 1.5 s relative to the H2|Air front propagation. A 100-cm² single cell with state-of-art flow field and commercial MEA is used. This ageing test is coupled with electrochemical characterization by voltammogram, CO / CO2 gas analysis at the cathode outlet and performance measurement by polarization curve to assess a clear insight into degradation mechanisms. Afterwards a sensitivity study is performed onto both plateau potential value (Efront) and its exposure time (tfront). Results validate that lowering cathodic potential during H2|Air front propagation by dummy load application and lowering its exposure time by accelerating H2 introduction are promising mitigation strategies. In further details, degradation mechanisms within CCL are more sensitive to potential parameter than the exposure time. Pt dissolution and redeposition are revealed predominant at lower potential range (1.0-1.4 V) whereas carbon support corrosion and Pt detachment are at higher potential range (1.4-1.6 V). Otherwise, this work shows also that SU-AST cycling with an Efront value equal to 1.0 and 1.2 V improves cell performance despite a loss of ECSA and carbon support. A beneficial CCL modification at the beginning of test is suspected, which could lead to an enhancement of future break-in protocol assuming optimize carbon support properties and thus cell performance at high current densities.