Kohta Yamada*, Satoru Hommura**, and Tetsuji Shimohira***
Asahi Glass Co., Ltd. has been developing a variety of perfluorinated ion exchange polymers (ionomers) in order to improve the performance, durability and robustness of membrane electrode assembly (MEA) for polymer electrolyte fuel cell (PEFC). In the present report, the effect of high oxygen permeable ionomers on MEA performance is described. At first, the oxygen permeability of some ionomers was determined by means of potential step chronoamperometry with a Pt microelectrode. It was confirmed that the newly developed ionomers had higher oxygen permeability compared with reference ionomer, and the permeability strongly depended on ionomer density. Then, the MEA performance was improved by using the high oxygen permeable ionomers in cathodes. These results also suggested that the high oxygen permeable ionomers would be available for the reduction of Pt loading in MEA cathodes.
*Asahi Glass Co.,Ltd., Technology General Division, New Product R&D Center, Planning & Administration Group **Asahi Glass Co.,Ltd., Technology General Division, New Product R&D Center, Business Incubation Group *** Asahi Glass Co.,Ltd.,Technology General Division, New Product R&D Center, New Product Group 2
−8− Res. Reports Asahi Glass Co., Ltd., 66（2016）
were determined by means of potential step
1. Introduction chronoamperometry (PSCA) using a Pt microdisk Significant cost reduction of MEA is required for the electrode and a dynamic hydrogen electrode widespread commercialization of PEFC systems for (DHE), which was similar technique written in the fuel cell vehicle (FCV) and residential use. Especially, published papers (7,8). The working electrode was a the reduction of Pt loading in MEA cathode is thought Pt microdisk electrode of the diameter of 50μm which to be effective for the purpose. At the cathode, the was sealed in a glass rod. The counter electrode was oxygen reduction reaction (ORR) proceeds on a Pt a Pt wire of the diameter of 1mm and the length of electrocatalyst supported on carbon in a porous gas 50mm. A DHE, which was a pair of 300μm Pt wire diffusion electrode. Although the kinetics and in a glass rod, was used as a reference electrode. mechanism of the ORR have been investigated with a The electrode reactions for DHE were hydrogen various kinds of electrocatalyst materials and evolution and oxygen evolution on each electrode electrolytess (1-6), the Pt or Pt-alloy have been the best from water electrolysis (9). A small current of 3 μ electrocatalyst up to now. However, the ORR on Pt A was applied, and electrode side of hydrogen electrocatalyst is very irreversible, and in addition the evolution was used as a reference electrode. In mass transport of reactive species dominates the ORR order to obtain good contact between electrodes at high current density. Therefore, as the cathode Pt and ionomer samples, each electrode was pressed loading is reduced, flux of oxygen near the Pt on the sample which was the membrane of the electrocatalyst surface should be increased, that is, the thickness of 200μm prepared by casting the Pt specific current density should be improved. In the ionomer solutions. present report, the new perfluorinated ion exchange The measurement cell, which was similar to polymers having higher oxygen permeability (HOP that in the literature (7), was placed in an electric ionomers) than the reference ionomer were evaluated. oven, and the temperature was maintained at 80℃ Additionally, the effect of the HOP ionomers on MEA under an atmospheric pressure. The humidified performance was investigated. air was flowed into the cell, and relative humidity (RH) was controlled between 30% and 80%. As for electrochemical measurements, cyclic 2. Experimental voltammetry and PSCA were carried out by using a potentiostat ALS Model 720D. Prior to 2.1 Ionomers tested electrochemical measurements all the electrodes At first, new perfluorinated HOP ionomers were were pretreated by scanning repeatedly between prepared for the present study. The characteristic 0.0 and 1.4V vs RHE in 0.1 mol·l-1 H2SO4 at a rate of features of the ionomers showed the lower density 0.2 V·sec-1. In order to in-situ u clean up and activate compared with the reference ionomer. Table 1 Pt microdisk surface in the cell, the potential of shows the density and the ion exchange capacity the working electrode was repeatedly scanned (IEC) of the HOP ionomers. Here, the density was between 0.05 and 1.5 V vs. DHE at 0.1 V·sec -1 measured as a precursor form without hydrolysis. before each electrochemical measurements. The IEC of these ionomers were fixed at 1.5 From the PSCA measurements, mass transport meq.·g-1. In the case of the measurements of oxygen parameters of oxygen in ionomers (C, D and CD D) permeability and MEA performance, these ionomers were obtained (C : solubility, D : diffusion coefficient, were used after hydrolysis treatment to H-form. and CDD : permeability). The potential of working electrode was stepped from 1.2 to 0.5 V vs. DHE, Table 1 Properties of ionomers tested. at which ORR is under diffusion limiting conditions, and the transient currents were monitored in the period of 20 sec. In general, the transient current for PSCA using a microdisk electrode is described by the following equation (10,11):
2.2 Oxygen permeability and solubility in Here, n, F, A and r are the number of electrons ionomers transferred (4 for ORR on Pt), the Faraday constant, The oxygen permeability and solubility the geometrical area of the microdisk electrode, and
−9− 旭硝子研究報告 66（2016）
the radius of the microdisk, respectively. The mass
transport parameters were calculated from the slope 3. Results and Discussion and the intercept of the I vs. t -1/2 plots. 3.1 Oxygen Permeability and Solubility in 2.3 Evaluation of MEAs Ionomers In order to investigate the effect of the HOP Figure 1 shows the oxygen permeability of the ionomers on MEA performance, MEAs were ionomers at 80℃ in air from 30% to 80% RH. The fabricated as follows. Firstly, catalyst inks were permeability was around the order of 10-12 mol·cm-1·sec-1, prepared by mixing Pt/C catalyst (TEC10EA50E, and increased with raising the relative humidity. TKK), solvent and the ionomer listed in Table 1, These values and RH dependence were similar to so that the weight ratio of the ionomer content the earlier report (8). As a result, it was found that was 0.8 relative to the carbon support (I/C=0.8). the permeability strongly depended on the kinds Secondly, the cathode catalyst layers of Pt loading of ionomer, and increased with the reduction of the of 0.2 mg·cm-2 were obtained by casting and drying ionomer density. The permeability of HOP-2 ionomer the inks on a transfer sheet. Next, the cathode having the density of 1.91 showed ca. twice higher catalyst layers were decal-transferred on a CD values compared with that of the reference perfluorosulfonic ion exchange membrane with an ionomer. Figure 2 shows the oxygen solubility in anode catalyst layer. All the anode catalyst layer ionomers obtained from the analysis using equation was consisted of Pt/C catalyst (TEC10EA50E, (1). It was clear that the oxygen solubility increased TKK) and perfluorinated ion exchange ionomer with decreasing the ionomer density. These results (Flemion ®) of I/C=1.0. The Pt loading of the indicated that the oxygen dissolution and transport anode catalyst layer was kept at 0.05 mg·cm -2. in ionomers strongly depended on the ionomer Finally, MEAs were assembled with gas diffusion density, which was thought to be related to the free layers (GDL) on both sides of the catalyst-coated volume of ionomers. membranes. The GDLs used in this study were non-woven type carbon fiber substrates with microporous layer. The geometric area of the MEAs was 6.25 cm2. The IV performance of the MEAs was measured under the condition of H2/air of 0.75/2.0 L·min-1 of gas flow rate, at a cell temperature of 80 ℃ , 40% RH and 150 kPa-abs. In addition, cyclic voltammetry measurements were also carried out under a similar condition of H2/air of 0.3/0.3 L·min-1 of gas flow rate, at a cell temperature of 80℃ , 40% RH and 150 kPa-abs. The potential scan rate of 20 mV·sec-1 was used.
Fig.1 Oxygen permeability CD of ionomers as a function of
2.4 Evaluation of Low Pt Loading MEAs relative humidity at 80℃ in air. The low Pt loading MEAs were fabricated by the same method described above, except for the followings. The Pt/Co alloy catalyst (TEC36F62, TKK) was used for the cathode ink, and the Pt loading of the cathode catalyst layers was reduced to 0.10 and 0.15 from 0.2 mg·cm-2. The weight ratio of the ionomer content to the carbon support was 1.0 (I/C=1.0). The geometric area of the MEAs was 25 cm2. The IV performance of the MEAs was obtained under the condition of H 2/Air of 1.43/2.0 stoichiometry, at a cell temperature of 80 ℃ , 20% RH and 150 kPa-abs.
Fig.2 Oxygen solubility C of ionomers as a function of
ionomer density at 80℃ in air.
−10− Res. Reports Asahi Glass Co., Ltd., 66（2016）
3.2 Evaluation of MEAs
Next, the effect of the HOP ionomers on MEA performance was investigated. Figure 3 shows the effect of ionomer in cathode on MEA performance at 80℃, 40% RH. The cell voltage and the IR free voltage depended on the ionomer, and was drastically improved with increasing the oxygen permeability of ionomer in the entire range of current density. In Fig.4, the effect of ionomer on cyclic voltammograms under similar condition to the measurement of MEA performance is shown. The oxygen reduction current was clearly observed, and the absolute value of the current was increased with the oxygen permeability of ionomer. Therefore, one of the reasons for the improvement of MEA Fig.5 Cell voltage at 1.6 A·cm-2 as a function of Pt loading at 80℃ , 20% RH, 150 kPa-abs, H2/air=1.43/2.0 stoich. performance by using HOP ionomers was thought to be the low mass-transfer loss of oxygen near Pt 3.3 Evaluation of Low Pt Loading MEAs surface in cathodes, although the thermodynamic Figure 5 shows the cell voltage at 1.6 A·cm-2 as a potential gain was not corrected yet. function of Pt loading. As shown in the figure, the lower cell voltage was obtained in the case of MEA of low Pt loading. Then, it was found that the cell voltage of MEA assembled with the catalyst layer of Pt loading of 0.1 mg·cm-2 and HOP-2 ionomer was almost similar to that with Pt loading of 0.2mg·cm-2 and the reference ionomer. This result suggested that the HOP ionomer would be available for the reduction of Pt loading in MEA cathodes, which is required for widespread commercialization of PEFC system.
4. Conclusion The mass transport parameters of oxygen in the newly synthesized perfluorinated ionomers were evaluated by means of electrochemical techniques. Fig.3 Effect of ionomer on MEA performance at 80℃ , As a result it was confirmed that the permeability 40% RH, 150 kPa-abs, H2/air=0.75/2.5 L·min-1, strongly depended on the ionomer density. Then, solid: cell voltage, broken: IR free voltage. the IV performance was found to be improved by using the HOP ionomers in MEA cathodes. These results indicated that the high oxygen permeable ionomer would be available for the reduction of Pt loading in MEA cathodes.
Fig.4 Effect of ionomer on cyclic voltammograms at 80℃ ,
̶ References̶ (1) K. Kinoshita, Electrochemical Oxygen Technology, p. 19, The Electrochemical Society Series 1117, Pennington, NJ (1992). (2) H. R. Kunz and G. A. Gruver, J. Electrochem. Soc., 122, p.1279 (1975). (3) K. Kinoshita, J. Electrochem. Soc., 137, p845 (1990). (4) A. J. Appleby, Catal. Rev., 4, p.211 (1970). (5) V. Jalan and E. J. Taylor, J. Electrochem. Soc., 130, p.2299 (1983). (6) M. Rao, A. Damjanovic and J. M. Bockris, J. Chem. Phys., 67, p.2508 (1963). (7) K. Ota, Y. Inoue, N. Motohira and N. Kamiya, J. New Mat. Electrochem. Systems, 3, p.193 (2000). (8) Y. Takamura, E. Nakashima, H. Yamada, A. Tasaka and N. Inaba, ECS Trans., 16(2), p.881 (2008). (9) A. Partharathy, S. Srinivasan, A. J. Appleby and C. R. Mar- tin, J. Electrochem. Soc., 139, p.2530 (1992). (10) C. P. Winlove, K. H. Parker and R. Oxenham, J. Electroanal. Chem., 170, p.293 (1984). (11) J. Maruyama, M. Inaba, T. Morita and Z. Ogumi, J. Electro- chem. Soc., 504, p.208 (2001).
