Bing Energy Reduced Pt Membrane Electrode Assemblies

Abstract

We have explored and demonstrated a novel nano-engineered electrode technology to meet the DOE’s targets for platinum group metal utilization and electrode durability for proton exchange membrane fuel cells. Our new electrode technology has made three major technical improvements of (1) the intrinsic electrocatalytic activity and stability of the catalystic electrodes; (2) the utilization of these materials using nano-structured and functionalized supports; and (3) the stability of the electrocatalyst/support system by strengthening the electrocatalyst-support interaction and by using corrosion resistant supports. A double-layered buckypaper (DLBP) was prepared by filtering the carbon nanotube (CNT)/carbon nanofiber (CNF) mixture and the CNF suspension sequentially under full vacuum. CNTs of ~10 nm in diameter and 10 mm in length and CNFs of 100-200 nm in diameter and 30-100 μm long. Platinum (Pt) nanoparticles were deposited onto the buckypaper using a pulse electrodeposition technique. As shown in Fig. 1(a), large-sized CNFs entangled randomly forming a highly porous sublayer with a porosity of 90.8% and an average pore size of 85 nm, while fewer and smaller pores were formed in the CNT/CNF sublayer by adding 25 wt.% smaller CNTs. After depositing Pt on the buckypaper by electrochemical deposition, an energy-dispersive X-ray spectrometer analysis showed an obvious gradient distribution of Pt (Fig. 1(b)); over 70% Pt was distributed in the 7 mm-thick CNT/CNF sublayer. A large amount of Pt deposited on the surface of the CNT/CNF sublayer (Fig. 1(c)), while only a small amount deposited on the surface of the CNF sublayer (Fig. 1(d)). This could provide an indication of how Pt was distributed inside the buckypaper since the Pt distribution was quite uniform within each sublayer. The pictures of surface morphology showed the Pt preferred to grow on the surface of CNT instead of CNF. Although, the mechanism of selective deposition of Pt on CNT/CNF buckypaper is not fully understood, the most possible reasons we suppose are: (1) CNT has much higher surface area and more surface defects resulting in more anchor sites for Pt nucleation; and (2) a higher deposition current on CNTs because of their higher conductivity. By applying this double-layered buckypaper as a cathode catalyst layer, the MEA exhibited excellent power performance with a relative low Pt loading. The rated powers of 1.5 W/cm2 and 0.8 W/cm2 at 0.65 V were achieved with a cathode Pt loading of 0.2 mgPt/cm2 (Fig. 2) for oxygen and air, respectively. The Pt utilization was better than the current state-of-the-art value (>0.4 mgPt/cm2) achieved by the conventional Pt/C catalys at a comparable cell (0.8–1.0W/cm2) output. The CNT/CNF buckypaper-based Pt catalyst has also shown good durability under an accelerated degradation test in a mimic cathode environment. This was due to the high corrosion resistance of CNF and its high graphitization degree. Hence, we evaluated the durability of catalyst support for Pt/DLBP in a MEA per US DOE’s suggested protocol. Figure 1