(419b) Steady State Temperature Dependent Carbon Oxidation Kinetics for Proton Exchange Membrane Fuel Cells

Vehicle applications
for proton exchange membrane (PEM) fuel cells require low cost, durable
materials to compete with incumbent propulsion technology.  As a result, recent
research has focused on reducing the quantity of precious metal needed to
catalyze the oxygen reduction reaction (ORR) in the cathode electrode of the
fuel cell.  Some of the most promising low-loaded catalysts have been developed
on high surface area carbon (HSC) supports, which enable fine dispersion of the
precious metal catalyst, leading to higher mass activities.  Unfortunately, HSC
supports (such as Ketjen) are also more corrosion susceptible than previously
used catalyst supports, such as Vulcan XC-72.  Corrosion of the HSC support can
promote catalyst sintering and increase electrode transport resistance.  Consequently,
vehicle systems are being developed to mitigate the most damaging events to the
HSC support.  One such event is cathode air storage during vehicle off times,
where data has indicated that the cathode potential can reach about 0.96-1.0 V,
well above the equilibrium corrosion potential of 0.207 V.  As a result, it is
critical to understand how much carbon loss can result from an air storage
event.  This work is focused on developing an empirical kinetic model for carbon
loss under steady state high potential (> 1.0 V) conditions at various
operating temperatures.

Low-loaded MEAs (0.2 mg_Pt/cm²
30% Pt-alloy catalyst on HSC support) were exposed to steady-state high
potential (1.0-1.4 V) operation at various temperatures (30°C,
55°C,
80°C)
to induce HSC support oxidation to carbon dioxide (CO₂).  The
cathode outlet gas was analyzed using an in-line nondispersive infrared (NDIR) carbon
dioxide detector at 1 Hz to establish a corrosion current vs. time profile.  A
form of the empirical equation for air storage corrosion current as a function
of potential, temperature, and time was proposed, and an attempt was made to
fit three coefficients to the collected data using multivariate linear
regression.  These kinetics were then used to perform a sensitivity analysis of
number of air storage events versus duration of air storage events, comparing
total predicted carbon loss from air storage over the life of each vehicle
usage profile.

Finally, the
characteristic decay of CO₂ concentration over time during the
steady state potential hold was used to hypothesize the possible impact of surface
coverage (both catalyst and carbon) on carbon oxidation kinetics.  Based on
this hypothesis, future potential cycling testing to measure carbon oxidation
is proposed.