(665d) Unraveling the Effect of Electrochemical Pre-Treatment on the Structure and Performance of the Electrocatalysts

Activation, break-in, or pre-treatment protocols are electrochemical techniques applied in energy-storage or -conversion devices, like fuel cells, before regular operation. While these protocols are intended to maximize the utilization of electrocatalysts, their effect on the intrinsic physical properties and the performance of the catalyst remain poorly understood. Here, we explored the role of electrochemical pre-treatment protocols by applying a design-of-experiments-based (DoE) approach to investigate the electrocatalytic oxygen reduction reaction (ORR) on a commercial Pt/C catalyst. Different pre-treatment protocols were generated based on a central composite design for five levels of five electrochemical parameters: upper potential (UPL), potential depth, sweep rate, number of cycles, and hold time at potential extrema. The experiments were performed in a flow cell combined with an inductively-coupled plasma mass spectrometer (ICP-MS). Our customized online ICP-MS setup provided simultaneous and real-time information about the electrochemical performance (mass-transfer limited current densities) and catalyst dissolution (as low as 0.1 ppb). The results indicate that applying high UPLs (~2V) during pre-treatment exhibited lower dissolution during the pre-treatment cycle than protocols with lower UPLs. However, the onset potential decreased ~5-fold during the electrochemical load cycle. On the contrary, the protocols with intermediate values of UPLs exhibited higher dissolution during pre-treatment and high onset potential during the electrochemical load cycle. The increased dissolution in pre-treatment occurred primarily due to increased dissolution while sweeping from oxidizing to reducing currents. Such insights, combined with various ex-situ structural characterization techniques, enabled us to hypothesize different physical mechanisms occurring during pre-treatment that govern the structure and final performance of the electrocatalyst. Finally, we propose a roadmap to integrate such findings to develop a causal learning model. Thus, combining DoE, online-ICP MS, and ex-situ characterization, we provide a powerful approach for gaining a mechanistic understanding of pre-treatment protocols.