Oxygen Reduction Reaction Activity and Durability of Pt-Ni Catalysts Synthesized By a Continuous Flow Reactor

Abstract

In recent years, polymer electrolyte membrane fuel cell (PEMFC) technology has attracted much attention and enabled the successful commercialization of fuel cell electric vehicles (FCEVs). However, the continuation of FCEV deployment still faces various hurdles, of which one major challenge is related to the high cost of platinum-based catalysts.[1] Substantial efforts have been dedicated for improving catalyst structure and composition to enhance their performance and thus lower the loadings for cost reduction. For example, shape-controlled catalysts such as octahedral PtNi and PtCo alloy nanoparticles showed much higher activity for oxygen reduction reaction (ORR) than the conventional Pt/C catalyst. [2-5] However, those advanced catalysts were typically prepared by batch synthesis with quantities often as little as a few milligrams, making it difficult for validation and further development of membrane electrode assembly (MEA) for commercial application. Attempt in increasing the catalyst amount by multiple batches can be time-consuming and sometimes challenging to maintain the consistency.We recently designed a flow reactor (FR) to explore the continuous synthesis of Pt alloy nanoparticles (Fig. 1.) under elevated temperature and pressure (up to 350 °C and 1000 psi). It has several advantages including fast heat-exchange, fast mixing, uniform concentration and uniform temperature as compared to batch synthesis. These features made it possible to accurately control the reaction parameters to produce monodisperse Pt alloy catalysts up to 10 grams per day with high reproducibility. In this study, systematic investigation was conducted on the FR synthetic conditions for highly active catalyst. We were able to control the size and composition of resulting PtNi nanoparticles to achieve much higher ORR activity than commercial catalysts by optimizing synthesis conditions. As shown in Fig. 2 and Fig. 3, octahedral PtNi alloy catalysts synthesized by the FR showed much higher mass activity and specific activity compared to commercial benchmark. Attempts were also made to use machine learning tools to assist experimental design and data analysis for process optimization. It is expected that the combination of the latest development of data science and the inherent features of such flow reactors can substantially accelerate future FC catalyst R&D. Figure 1