Polymer electrode fuel cells (PEFCs) have various positive aspects for improving the worldwide environmental problems. But PEFCs have some issues that restrict commercialization of PEFCs. The biggest issue is its high cost because of using platinum, expensive metal, as catalyst. So, efficient utilization of expensive platinum is necessary to reduce the cost. The resent study reported that graphene has not only high electrical conductivity but also enough proton conductivity (1). It was also reported by the authors that the oxygen transport resistance in the cathode catalyst layer (CL) is dominant at interface of ionomer (2). In this study, we introduced graphene nano platelets (GNPs) as support material for platinum instead of conventional carbon. This idea has possibility that we can reduce the amount of ionomer and the oxygen transport resistance in the CL drastically, resulting in more effective utilization of platinum.

In this study, cathode CLs with various ionomer-carbon ratio (I/Cs) were fabricated using self-made Pt-supported GNPs by decal transfer method. The structure and the performance of fabricated CL were evaluated by I-V measurement, SEM observation, and N₂ physisorption analysis. These were also compared with fabricated CLs using commercial Pt-supported Ketjen Blacks. At the anode side, the CL with the Pt loading 0.20mg/cm² and I/C 0.8 was used. Operation temperature was 80℃ and relative humidity (RH) of supply gases was 80%.

First, the effects of I/C on the cell performance were investigated. Measured I-V curves for the graphene-based CL with I/C 0.4 was better than that with I/C 0.8. This tendency is different from that for the Ketjen Black based CL, suggesting the possibility that ionomer can be reduced.

However, the performance of the graphene-based CLs was lower than the Ketjen Black based CL. This was estimated to be caused by the poor pore structure in the graphene-based CLs. The N₂ physisorption analysis showed that the graphene-based CL has few secondary pores around 100nm and the porosity with I/C 0.4 is about half of that of I/C 0.8 Ketjen Black based CL. On the other hand, the CL porosities were also estimated from the used component volumes and the thickness measured by SEM images, and the porosities for the two different CL were similar. These indicate that there are a lot of pores that oxygen cannot permeate enough in the I/C 0.4 graphene-based CL. Then, we tried to evaluate the I/C 0.0 graphene-based CL. Although it is difficult to fabricate CL with I/C 0.0, more effective pores for oxygen paths are expected with enough proton conductivity by graphene itself. Figure 1 shows the results with the I/C 0.0 graphene-based CL and the I/C 0.8 Ketjen Black CL. It can be confirmed that the graphene-based CL with Pt loading only 0.044mg/cm² achieves high performance similar to Ketjen Black based CL with Pt loading 0.21mg/cm². This indicates that graphene could play a significant role in large reduction of platinum in PEFC CLs.

References

(1) S. Hu, et.al, Nature, 516, 227 (2014).

(2) T. Hayashi, et.al, ECS Trans., 75(14), 373 (2016).