A 20-cell self-humidifying fuel cell stack containing two types of MEAs was
assembled and aged by a 1000-hour durability test. To rapidly and effectively
analyze the primary degradation, the polarization change curve is introduced. As
the different failure modes have a unique spectrum in the polarization change
curve, it can be regarded as the fingerprint of a special degradation mode for
repaid analysis. By means of this method, the main failure mode of two-type MEAs
was clearly distinguished: one was attributed to the pinhole formation at the
hydrogen outlet, and another was caused by catalyst degradation only, as
verified by infrared imaging. The two distinct degradation phases were also
classified: (i)conditioning phase, featuring with high decay rate, caused by
repaid ECSA change from particle size growth of catalyst. (ii) performance phase
with minor voltage loss at long test duration, but with RH cycling behind, as in
MEA1. Then, an effective H2-pumping recovery is conducted, and MEA
performance rejuvenated to the level of phase transition time (218h). It
indicated the voltage loss at the performance phase is reversible, while the
conditioning phase is not as expected. Therefore, there are few perquisites for
the durability test, including proper selection of current density point as a
baseline, test duration beyond the conditioning time, and effective recovery
procedure; otherwise, the lifetime of the stack would be extremely
underestimated. As massive voltage loss is from catalyst degradation in the
conditioning phase for all MEAs and featuring irreversible, more understanding
about catalyst degradation at the conditioning phase is suggested for further
durability improvement.