Optimization of Cathode Catalyst Layer of PEFC Using Silk-Derived Activated Carbon By 2-Step Mixing Method Introduction

Abstract

IntroductionFuel cell is expected to be used as one of the major power sources in the near future, from the view point of high energy efficient and low environmental impact. Among the fuel cells, polymer electrolyte fuel cell (PEFC) is beginning to be put into practical use as residential and transportation power generation system. However, there are still some problems to be overcome for the spread of PEFC. One of them is the high cost, due to the use of platinum (Pt). Pt is used as cathode catalyst for reduction of overpotential in most PEFCs. Since Pt is a metal of low abundance and high cost, alternative catalyst is required. Many oxides, nitride and carbon materials have been investigated as non-precious metal catalysts. However, most of them show low oxygen reduction reaction (ORR) activity, low conductivity or low durability. We have reported the performance of silk derived activated carbon (SAC) as a non-precious metal catalyst for the PEFC cathode1,2). Compared with Pt, SAC has lower ORR activity but it can be fabricated cheaper. Therefore, the optimum structure of the catalyst layer using SAC is expected to be different from that using Pt. In this study, we prepared cathode catalyst ink of SAC using different solvent, and investigated the effect of mixing ratio of the solvent and the effect of 2-step mixing to the performance, which is necessary to obtain information to optimize the electrode structure for non-Pt cathode for PEFC. ExperimentalSAC was prepared using silk fibroin as starting material after removing sericin from silk fiber. Silk fibroin was primary carbonized under nitrogen atmosphere at 500 ˚C. The obtained carbon was ball milled and then acid washed. Then it was heated under nitrogen atmosphere for secondary carbonization at 900 ˚C.Membrane electrode assembly (MEA) was prepared using SAC as cathode catalyst. Cathode catalyst layer was prepared by dropping the dispersion ink on the carbon paper. Butyl acetate (BA) and/or n-propanol (NP) were used as solvent. Dispersion ink was prepared by two methods, 1-step mixing and 2-step mixing. In 1-step mixing, carbon black (CB) and SAC were mixed in the solvent simultaneously. On the other hand, in 2-step mixing, CB and SAC were mixed separately. Catalyst layer was observed by scanning electron microscope (SEM) and energy dispersive X-ray spectrometer (EDX). Terminal voltage vs. current density was measured at 80 ˚C under ambient pressure, using hydrogen and oxygen for anode and cathode, respectively. Both gases were humidified by a vaporizer. Results and discussionFigure 1 shows polarization curves of MEA prepared by two types of solvent. As can be seen in the figure, in the low current density region, the cell performance of MEA prepared by using mixture of BA and NP as the solvent showed higher performance than that prepared by using only BA. On the other hand, the effect of the solvent to the performance was opposite in the high current density region. It is reported that different solvent affects the structure of the ionomer, therefore, the different effect observed in different current density region is assumed to be caused by the structure of the ionomer in the electrode.Figure 2 shows polarization curves of MEA prepared by1-step mixing and 2-step mixing. The maximum power density increased by 2-step mixing. The improvement performance by 2-step mixing is assumed to be caused by the optimized electrode structure, which was achieved by coating SAC with the ionomer during the first step, and by building up the continuous network of ionomer during the second step.1) T. Iwazaki, R. Obinata, W. Sugimoto and Y. Takasu, Electrochem. Comm., 11, 376 (2009). 2) H. Fukunaga, R. Shishido, T. Takatsuka and Y. Takasu, Innovative Materials for Processes in Energy Systems 2010. B. B. Saha et al., eds. Research Publishing, 142 (2011). Figure 1