Polymer electrolyte fuel cells (PEFCs) have various positive aspects for achieving carbon neutrality because carbon dioxide is not emitted during operation. However, there are several issues for practical use of PEFCs. One of the problems is its high cost due to the use of Platinum (Pt) for catalyst. Our previous study tried to reduce the amount of Pt in PEFC catalyst layer (CL) and realized low Pt-loading CL by using graphene as Pt support material (1). The graphene CL did not contain ionomer which prevent oxygen transport to the Pt surface. Therefore, we considered that graphene CL realizes smooth oxygen transport. However it was found that the proton transport resistance of graphene CL was about ten times that of CL containing ionomer (2). Since the proton transport resistance depended on the CL thickness, present study employed an electrospray (ES) method to fabricate thin CL. If the proton resistance in graphene CL is reduced, Pt can be used more efficiently.

In this study, graphene CLs with various Pt-loading (mg/cm²) were fabricated by decal transfer method, gas diffusion electrode (GDE) method or ES method. The graphene CL was used in the cathode side. The performance and the proton resistance of fabricated CL were respectively evaluated by I-V measurement and impedance measurement. These were also compared with fabricated CLs using commercial Pt-supported Ketjen Blacks (KB/Pt CL). At the anode side, the CL with the Pt loading 0.20mg/cm² and I/C 0.8 was used. Operation temperature was 80℃, relative humidity (RH) of supply gases was 80%, the flow rate of cathode gas was 2000 sccm with 1% oxygen concentration, and the flow rate of anode gas is 100 sccm.

Relationship between fabrication method and the I-V performance was investigated. Figure 1 shows the I-V performance of KB/Pt CL with different fabrication method at 1% oxygen concentration. The CLs with the same Pt-loading (0.047mg/cm²) but different fabrication methods (transfer, ES) show similar I-V performance. This result indicates that fabrication method does not affect the I-V performance in case of KB/Pt CL.

Figure 2 shows the I-V performance of graphene CL with different fabrication method at 1% oxygen concentration. The performance of graphene CL fabricated by ES method was higher than that with GDE method despite similar Pt-loading and proton resistance (Figure 3). In case of graphene CL, the structure is considered to depend on the fabrication method. The CL structure fabricated by ES method can improves water and oxygen transport in case of graphene CL.

Figure 3 shows the proton transport resistance of the fabricated catalyst layers. The proton resistance is reduced in the smaller amount of Pt-loading CL. Additionally, the Pt-loading in graphene CL was proportional to the thickness of the CL because the weight of Pt relative to the that of graphene was constant (17.2wt%). Therefore, it was suggested that the proton resistance was reduced by fabricating a thinner catalyst layer.

Comparing the I-V performance of KB/Pt CL 0.047mg/cm² (Figure 1) and graphene CL 0.01mg/cm² (Figure 2), the graphene CL shows high performance even though the Pt weight was about 1/5. These results indicate that fabrication of thin graphene CL with ES method enables to use Pt effectively. Higher performance of graphene CL was also confirmed under higher oxygen concentration condition. Figure 4 shows the I-V performance of graphene CL with ES (Pt-loading: 0.01mg/cm²) and KB/Pt CL with transfer method (Pt-loading: 0.2mg/cm²) at 21% oxygen concentration. Both of limiting current density were over 1 A/cm². Even though the difference in Pt-loading was 1/20, the difference in I-V performance was small.

References

(1) K.Yada, et.al, ECS Abstract, #101A-1389, (2019).

(2) M.Okano, et al., The 25^th National Symposium on Power and Energy, C223(2021)(in Japanase).