Development of Non-Spherical Platinum Catalyst with Functionalized Carbon Supports for Proton Exchange Membrane Fuel Cells
Loading...
Date
2023-04-21
Authors
Lim, Mark
Journal Title
Journal ISSN
Volume Title
Publisher
University of Waterloo
Abstract
Energy conversion devices capable of zero-emission operation are needed to limit the impacts of anthropogenic climate change. Proton exchange membrane (PEM) fuel cells are well suited to address this need, as they can produce electricity directly from the chemical energy stored within green hydrogen and oxygen, emitting only water and heat. A major component of their cost is platinum (Pt) based catalyst, which is needed to catalyze the oxygen reduction reaction (ORR). Therefore, the performance and durability of Pt-based catalysts must be improved to drive down the cost of PEM fuel cells. One strategy is to adopt non-spherical morphologies, which can increase the active surface area and inherent ORR activity compared to spherical Pt nanoparticles (NPs). Another strategy is to disperse the NPs on an optimally designed carbon support, forming a platinum/carbon (Pt/C) catalyst. The carbon support can be functionalized to introduce binding sites and promote dispersion of Pt NPs. While research has proven both these strategies effective, their benefits can only be fully realized if the Pt/C catalyst is synthesized in a facile and scalable manner while retaining its catalytic properties. As such, synthesis methods that use a one-pot approach and avoid difficult-to-remove surfactants are advantageous. In addition, practical applications necessitate the catalyst to perform well not only electrochemically, but also in full-cell tests. To date, few studies have emphasized the importance of facile/scalable synthesis and full-cell performance for shape-controlled Pt catalysts and functionalized carbon supports.
In this study, a surfactant-free one-pot method is developed to synthesize non-spherical Pt NPs on Ketjen Black carbon, and the catalysts are characterized using physical and electrochemical methods. First, the catalyst is synthesized in small batches on three varieties of Ketjen Black: non-functionalized (Pt/KB), oxidation-treated (Pt/KB-O), and nitrogen-doped (Pt/KB-N). Then, the catalysts are synthesized on a 10x larger scale to evaluate the impact of scaling up the synthesis. Surface area measurement and X-ray photoelectron spectroscopy measurements are done on the functionalized carbons, while transmission electron microscope imaging, X-ray diffraction, and thermogravimetric analysis are used to characterize the Pt/C catalysts. Electrochemical characterization on a rotating disk electrode setup is done to measure the catalysts' ORR activity. Accelerated stress tests are run by cycling the potential between 0.5-1 V and 1-1.5 V to evaluate their durability against catalyst degradation and carbon corrosion, respectively.
Physical tests confirm that both oxidation treatment and nitrogen doping affect the carbon surface chemistry and microstructure, which in turn affect the properties of the supported Pt catalysts. For both Pt/KB-O and Pt/KB-N, the functional groups on the carbon serve as binding sites, preventing the NPs from agglomerating. However, the carbon affects the NP size distribution, with Pt/KB-O containing larger NPs compared to Pt/KB and Pt/KB-N. When the synthesis is scaled up, Pt/KB changes greatly with its small NPs being replaced by nanowires or nanorods, whereas Pt/KB-O and Pt/KB-N remain visually similar between their small- and large-scale versions. Therefore, functional groups on the carbon foster a more predictable or scalable catalyst morphology when using this one-pot synthesis method.
Electrochemical tests reveal both similarities and differences between the catalysts on different carbon supports. Pt/KB-O achieves only similar or slightly lower ORR activity as Pt/KB, possibly because of poisoning from the sulfonate-based ionomer used in the test, and Pt/KB-N achieves similar or slightly higher ORR activity. All the catalysts are similarly durable when cycled between 0.5-1 V, showing that carbon functionalization does little to prevent Ostwald ripening. However, Pt/KB-O is the most durable when cycled between 1-1.5 V, showing that oxygen functional groups change the susceptibility to carbon corrosion or NP detachment. Unexpectedly, all three catalysts achieve higher ORR activity when the synthesis is scaled up, attributed to the high surface area and activity of Pt nanorods. These results will help inform the implementation of shape-controlled Pt catalysts and functionalized carbon supports in large scale.
Description
Keywords
shape-controlled catalysts, carbon support functionalization, nitrogen doping, one-pot synthesis, ORR activity, accelerated stress tests, proton exchange membrane fuel cells