Fuel cells are actively being commercialized for large-scale applications, such as zero-pollution automated systems, but are held back by high manufacturing costs, expensive catalysts, and the fuel delivery cost. While significant cost reductions have been made, the membrane electrode assembly components alone account for as much as 43% of the total system cost of a 2020 auto system at 1,000 units/year and 23% at 500,000 units/year. Fundamental improvements to membrane electrode assemblies (MEA) are occurring rapidly in many ways, such as through reduced catalyst loading, alternative catalysts, novel electrodes, improved electrode fabrication, and better assembly processes. However, quantifying improvements can be challenging due to small-scale assembly practices that induce non-uniformities, or defects, into an MEA. Current manufacturing practices rely heavily on cell testing to indicate material issues, which leads to delayed correction of production issues. At the research scale, all variation needs to be controlled so the changing variables (e.g., loading) have their effects properly characterized. Mainstream Engineering has developed and evaluated an in-line optical scanner that can determine membrane thickness, identify defects, and measure catalyst loading and variability. The inspection system samples every location so that defects in the materials can be selectively removed prior to assembly. This technology can improve the repeatability of data on new technologies and enhance rapid prototyping.
A variety of materials have been examined including proton exchange membrane (17 – 250 µm thick, supported and unsupported), anion exchange membrane (30 – 50 µm thick, reinforced), hydrocarbon membrane (20 – 165 µm thick, unsupported), catalyst coated membrane (Pt and Ir, 0.1 – 0.4 mg/cm2), gas diffusion electrodes (Pt/C 0.1 – 0.4 Pt/cm2), and a variety of support/protective layers. Defects in these materials (from 5 – 500 µm in size) were induced and optical features determined. The impact of induced pinhole defects on MEA performance and lifetime were examined through initial testing and accelerated stress tests. These tests provide a baseline for what defects must be seen, where they impact performance, and how improved in-line quality control can reduce assembled component failures. Sensor development, capabilities, impact of defects on membrane and fuel cell performance, and cost benefits during manufacturing and testing will be presented.
