Evaluation of Humidity Dependent O ₂ Transport Resistance in Primary Pores of Mesoporous Carbon Support

Mesoporous carbon (MPC) is a highly promising catalyst support material for cathode catalyst layers of proton exchange membrane fuel cells (PEMFCs) because it enhances catalyst activity for oxygen reduction reaction (ORR) and oxygen transport property [1-3]. A primary challenge lies in designing an optimal MPC support structure, including particle size and pore size distribution, in order to improve performance of a membrane electrode assembly (MEA). A key issue is to achieve proton accessibility to catalyst metal particles under low relative humidity (R.H.) condition and robustness against water condensation in the primary pores [4-6]. In this study, O 2 transport resistance in the catalyst layer with an MPC support was experimentally evaluated at various humidity. We extracted the transport resistance component of the primary pore of the support and investigated the contribution of water condensation to it. We analyzed an apparent local O 2 transport resistance which can be obtained experimentally by cell evaluation using a quasi-equivalent circuit shown in Figure 1. Here, we assumed that the platinum particles on the outer surface of the MPC were covered by ionomer, and the others on the inner surface were uncovered. The apparent resistance was divided into the components of the ionomer permeation resistance ( R ion ) and the effective diffusion resistance in the primary pores ( R pore ). Parameters required to determine the primary pore component were evaluated experimentally, including the fraction of the catalyst surface area on the outer surface of the MPC ( θ ) and the ionomer permeation resistance. A Pt-Co alloy catalyst supported on mesoporous carbon was used as the test sample. Its specification is summarized in Table 1. The MEAs were prepared by a pulsed spray coating method directly on a membrane. Nafion D2020CS (Chemous) was employed as ionomer. The MEAs were prepared with varying catalyst loadings in the range of 0.02 - 0.2 mg Pt cm -2 to evaluate the local O 2 transport resistance from the catalyst-surface-area dependence of the O 2 resistance in the catalyst layer. The limiting current measurement was performed under a relative humidity of 40 - 90%R.H. and a total pressure of 110 - 200 kPa, respectively. The relative humidity dependence of the effective O 2 transport resistance in the primary pores of the MPC is shown in Figure 2. The resistance increases significantly at higher relative humidities, especially above 60%RH. This trend is qualitatively consistent with the water adsorption isotherm property of the catalyst shown in Figure 3. The resistance represents a substantial change greater than that expected from the decrease in the porosity of the carbon particle due to water accumulation. It is hypothesized that a large proportion of the platinum particles were distributed in small diameter pores where capillary condensation preferentially occurs at lower humidity. The quantitative validity of this will be discussed in a future work using an O 2 transport simulation under water condensation conditions [7]. In contrast to the pore diffusion, the ionomer permeation resistance decreases with increasing humidity. This trend is consistent with the previous report on a model experiment of ionomer thin films on platinum [8]. We can discuss the respective contributions of the inner and outer catalyst particles to the cell performance under diffusion-limiting conditions based on the transport resistance components. This method would be a beneficial tool not only for designing optimal porous support structures with desirable oxygen transport properties and humidification robustness, but also for investigating an optimal spatial distribution of platinum particles on the inner and outer surface of MPCs. References [1] V. Yarlagadda, M. Carpenter, T. Moylan, R. Kukreja, R. Koestner, W. Gu, L. Thompson, and A. Kongkanand, ACS Energy Lett. , 3 , 618 (2018). [2] Y. Kamitaka, T. Takeshita, and Y. Morimoto, Catalysts , 8 , 230 (2018). [3] N. Ramaswamy, W. Gu, J. Ziegelbauer, and S. Kumaraguru, J. Electrochem. Soc. , 167 , 064515 (2020). [4] A. Chowdhury, C. Radke, and A. Weber, J. Electrochem. Soc. , 170 , 114525 (2023). [5] V. Yarlagadda, N. Mellott, S. Kumaraguru, and N. Ramaswamy, ACS Appl. Mater. Interfaces, 15, 55669 (2023). [6] N. Kikkawa and M. Kimura, Langmuir , 40 , 1674 (2024). [7] C. Otic, S. Katayama, M. Arao, M. Matsumoto, H. Imai, and I. Kinefuchi, ACS Appl. Mater. Interfaces , 16 , 20375 (2024). [8] R. Jinnouchi, K. Kudo, K. Kodama, N. Kitano, T. Suzuki, S. Minami, K. Shinozaki, N. Hasegawa, and A. Shinohara, Nat. Commun. , 12 , 4956 (2021). Acknowledgement This presentation is based on results obtained from a project, JPNP20003, commissioned by the New Energy and Industrial Technology Development Organization (NEDO). We would like to acknowledge Cataler Corporation for providing the catalyst utilized in this study. Figure 1