The catalyst layer (CL) is one of critical component for determining the performance of proton exchange membrane fuel cells (PEMFCs), and the production process of CL is important issue (Figure 1). In order to form the CL, catalyst ink mixed with platinum-supported carbon (Pt/C), ionomer solution, and a solvent is necessary. When the catalyst ink is coated on the sheet and dried, the CL is formed. By transferring the catalyst layer to the membrane, catalyst coated membrane can be made [1]. The process from the initial characteristics of the catalyst ink to the drying process determines the structure of CL and dominates the characteristics of the fuel cell. Thus, quality control of the catalyst ink is important to form a high performance and uniform CL [2].

In the previous study, we have reported that oxidation reaction of alcohol, as catalyzed by platinum, resulted in degradation of the catalyst ink [3]. However, the impact of catalyst ink degradation on the catalyst layer formation process has not been clarified. In this study, the composition change of the catalyst ink was investigated in more detail using gas chromatography mass spectrometry (GC/MS). Furthermore, the catalyst layer was prepared from the catalyst ink, and the influence of the catalyst ink degradation on the formation of the micro- to millimeter scale structure of the catalyst layer was investigated.

Catalyst ink was prepared by using the mixture of water and 1-propanol (NPA) as a dispersant. We used platinum catalyst supported on carbon black and Nafion dispersion D2020 (Chemours). Pre-mixing of catalyst ink was performed using ultrasonicaton to break large Pt/C agglomerates into smaller units. As a main mixing process, a high-speed rotary type mixer Filmix (Primix) utilizing shearing force, specialized for producing nanometer slurry was used. The prepared ink was immediately transferred to a glass sample container and sealed tightly. After 20 days, the composition of the catalyst ink was analyzed using a GC/MS (JMS-700, JEOL, Tokyo, Japan).

In Figure 2, result of GC/MS indicates that new 4 compounds were present in the ink composition in addition to water and NPA. Production of propionaldehyde and propionic acid was reported in the previous study [3], but present result showed production of propyl propionate and propane, 1,1-dipropoxy-. It is considered that not only oxidation reaction of the NPA but also dehydration reaction of each compound was occurred as shown in Figure 3. Both physical properties indicate low solubility in water. Since the liquid phase of the catalyst ink is composed of hydrophilic components, these hydrophobic compounds might cause phase separation and affect the production process of CL.

Figure 4 shows the results of microscope observation of the CL made from the catalyst ink. Many cracks existed in the formed CL. It is considered that the hydrophobic compounds caused phase separation in the catalyst ink, and those compounds acted as rupture point in the CL formation process. As a comparison experiment, another catalyst layer was made from catalyst ink using ethanol (C₂H₆O) as a solvent. Even if ethanol follows the same reaction path as in Figure 3, no hydrophobic compounds will be produced. Thus, it is considered that cracks are hardly formed. Figure 5 shows the CL surface made from the catalyst ink using ethanol, and no crack was formed. These results indicate that when hydrophobic compounds are produced from solvent alcohol by the catalytic action of Pt in the catalyst ink, the formation process of the CL and the quality of the formed CL are significantly affected.

Acknowledgment:

This study is based on the results obtained from a project commissioned by the New Energy and Industrial Technology Development Organization (NEDO). The authors thank the Suzukakedai Materials Analysis Division for performing sample analysis using GC/MS.

References:

• Suzuki, ECS Transactions, 75(14), 423-434 (2016).
• J. Shin et al., J. Power Source 106, 146-152 (2002).
• Uemura et al., J. Electrochem. Soc., 165-3, F142 (2018).