Preparation and Performance Evaluation of Pt/SnO2/KB As Cathode Catalyst of PEMFC

Several technical issues must be addressed to allow widespread commercialization of proton exchange membrane fuel cells (PEMFCs). In particular, the electrode catalyst that has high durability and an acceptable low cost (reducing Pt amount) is required.

For this purpose, we prepared novel electrode supports which are modified Ketjen black with tin dioxide nanoparticles (SnO₂/KB) and reported oxygen reduction reaction (ORR) activity as well as the durability of the resultant catalyst (after Pt loading) [1]. In particular, the ORR activity and the durability were strongly depended on the loading amounts of SnO₂ as well as the heat treatment temperature [1]. In this presentation, the relation between the preparation condition of SnO₂ / KB and the catalytic performance of the resultant catalyst are reported.

The preparation detail of SnO₂/KB can be found in our previous reports [1]. In brief, KB (Ketjen Black EC-300j) was dispersed in 0.01 mol dm^-3 SnF₂ aqueous solution for 24 h or 72 h.  It was then dispersed in 0.01 mol dm^-3 H₃BO₃ aqueous solution for 24 h. Subsequent heat treatment was carried out at 500 ^oC under Ar atmosphere.

Loading of Pt nanoparticles on SnO2/KB was carried out by an impregnation method following drying under N₂ atmosphere at 120 ^oC [1]. The catalysts were characterized by XRD, TGA under air, N₂ adsorption and desorption measurements, FE-SEM observation, ICP measurements and CO pulse tests.

The prepared catalysts were dispersed ultrasonically in 2-propanol and were dropped on a glassy carbon disk. A thin Nafion film with a calculated thickness of 80 nm was then formed. A glass half-cell equipped with an RHE (reversible hydrogen electrode) and a Pt mesh counter electrode was used in this study. 0.1 mol dm^-3 HClO₄ aqueous solution was used as the electrolyte. ORR activity was evaluated by Koutecky-Levich plot analysis based on the rotating disk electrode measurements under O₂ atmosphere.

The potential pulse durability tests were conducted between 1.0 V and 0.6 V at 60 ^oC.  Electrochemical surface area (ECA) of Pt was estimated by cyclic voltammetry.

From the TGA results, the weight ratios of SnO₂ of SnO₂/KB were 39wt% after 24 h in SnF₂ solution and 35wt% after 2 h, respectively. Then, we hereafter called the catalyst prepared by 24 h-dispersing in SnF₂ solution as Pt/39-SnO2/KB-24 and another as Pt/35-SnO2/KB-72.

In FE-SEM images of Pt/39-SnO2/KB-24 and Pt/35-SnO2/KB-72, SnO₂ particles of ca. 10 nm in the diameter were found on the surface of KB. On the other hand, Pt particle size could be estimated as 2 - 3 nm, which corresponding to the results of CO pulse tests. In addition, the particle size is estimated as 1.8 nm for Pt/KB.

Table 1 shows the Pt particle size, the values of specific and mass activity for ORR and the ECA remaining after the durability test. The ORR activity is clearly increased by using SnO₂/KB as the support; especially, the mass activity was high although the Pt particle size was large. Furthermore, the ORR activity is likely to be improved by increasing the time for dispersing in SnF₂ aqueous solution. The durability is also improved by using SnO₂/KB as the support.

Then, we attempted to investigate the reasons for higher durability of the catalysts using SnO₂/KB based on the degradation mechanisms.

The stability of the supports (SnO₂/KB) was elucidated the increasing rate of pseudo-capacitance during the potential pulse durability tests. The increasing rates of KB and Pt/35-SnO₂/KB-72 were 12.5% and 8.5%, respectively, which shows that the durability of Pt/35-SnO₂/KB-72 is higher than that of KB. In addition, to elucidate chemical stability of Pt, the stability tests were also conducted in O₂ saturated HClO₄ aqueous solution at 60 ^oC for 72 h. As the results, the dissolution ratio of Pt/35-SnO₂/KB-72 was ca. 0.8%, which was almost half for the catalyst without SnO₂.

The migration of Pt particles was also one degradation factor. Then, we carried out identical location FE-SEM observation of the catalysts before and after heating at 300 ^oC under Ar atmosphere for 12 h [2]. The frequency for migration of Pt particles was found to be ca. 58% for Pt/KB. On the other hand, that of Pt/35-SnO₂/KB-72 was 42%, which suggests that migration of Pt particles were depressed by using SnO₂/KB as the support.

[1] T. Kinumoto, et al, ECS Trans., 64(3), 199 (2014), 58(1), 1259 (2013), 50(2), 1701 (2012) and Electrochemistry, 79, 334 (2011).

[2] T. Kinumoto, et al., Electrochemistry, 83(1), 12 (2015).