PNNL

Abstract

Bipolar plates represent a significant portion of the cost and weight of a proton exchange membrane (PEM) fuel cell stack. As a result, there has been significant interest in using low weight and cost materials, such as aluminum as a material for bipolar plates. Aluminum is advantageous and has properties that can potentially allow it to attain technical targets required for bipolar plates such as high electrical conductivity, low contact resistance, good formability, and flexural strength. However, aluminum performs poorly in the corrosive environment within the fuel cell stack, prohibiting it from being used as bipolar plate material by itself. To overcome the corrosion performance issue, there have been efforts towards developing coating processes to enable aluminum to attain corrosion targets. In previous efforts to develop a coating process for aluminum, only melt-based coating processes have been investigated which shows a common issue of the existence of pores that eventually allow corrosive media to contact the aluminum. In an effort to develop a means to allow aluminum as a viable bipolar plate material, an alternative approach using a solid phase process (diffusion bonding) for bonding a thin layer titanium foil to aluminum was investigated in this study. Diffusion bonding of titanium to aluminum has the potential to overcome the shortcomings of other attempts to coat aluminum for the PEM fuel cell bipolar plates. The microstructure, corrosion performance, mechanical properties and electrical resistance were investigated. The effect of addition of highly conductive gold particles after diffusion bonding, on contact resistance and corrosion behavior, was also studied and compared with diffusion bonded Ti to Al. The results indicate that diffusion bonded Ti to Al provides a viable alternative material combination for bipolar plate that avoids the micro pores and crevasses that are associated with fusion (melt) based coating processes.