Shape-Selected Fuel Cell Nanocatalysts for Reduced Dependence on Platinum

[ACSGCI]

Contributed by Luke T. Roling, Graduate Research Assistant, Department of Chemical & Biological Engineering, University of Wisconsin-Madison

One of the most exciting “green” energy production technologies is low-temperature fuel cells, particularly for portable applications. These fuel cells can replace traditional combustion engines in vehicles, instead using electrochemical reactions to power human travel. In addition to being highly efficient, fuel cells offer additional opportunities to be “green” if their fuels, such as hydrogen or methanol, are renewably sourced (e.g. from biomass conversion); they therefore avoid releasing fossil fuel carbon into the atmosphere and can help mitigate climate change.

A substantial obstacle to the widespread implementation of fuel cell-driven vehicles is the high cost of the fuel cell catalyst – particularly at the fuel cell cathode, at which expensive platinum catalysts are generally required to yield suitable activity for the oxygen reduction reaction (ORR). As a further complication, these catalysts are prone to dissolution in the fuel cell environment, further increasing the operational costs. This motivates the discovery of new catalysts with (1) reduced platinum content, to minimize cost; (2) higher catalytic activity than pure platinum; and (3) improved stability over long operating times. The ideal solution for energy production wouldn’t require the use of rare earth or platinum group elements. However, reducing the amount of platinum or palladium needed for a fuel cell while increasing its performance level is an important step in the right direction.

Recently, researchers developed a method for creating new catalysts based on platinum-nickel bimetallic nanoparticles. These nanoparticles undergo structural transformations to ultimately form hollow frames with platinum-skinned surfaces, and these resulting “nanoframes” demonstrated exceptional activity and stability for the ORR. However, the nanoframes might not expose the most desirable surface structures for optimal ORR activity. This work inspired the development of a new class of highly active and stable nanomaterials for the ORR that selectively expose desired crystal facets for optimizing the ORR catalytic performance.

Our team, composed of experts in inorganic synthesis as well as atomic-scale modeling from first-principles, has worked to develop such nanostructured electrocatalysts with high activity and stability for the ORR. These catalysts are synthesized by following specifically developed design protocols that selectively deposit layers of platinum atop palladium seeds to yield well-controlled nanocrystals in what is known as a core-shell structure. The geometry of the catalyst plays an important role in determining its surface atomic-scale structure, which in turn affects the catalytic activity. These seeds can be synthesized in a variety of well-defined shapes according to established protocols; the choice of seed determines the shape of the final electrocatalyst’s nanoparticles.

In our initial studies, our team demonstrated the syntheses of cubes and octahedra with palladium cores and ultrathin platinum shells of only a few atomic layers (Figure 1). We performed detailed electronic structure calculations to predict the ORR activity of these catalysts, identifying that platinum is compressed slightly when deposited atop the palladium cores. This compression slightly destabilizes ORR chemical intermediates on the surface, which improves the activity of these surfaces. Our calculations offered predictions of the most active surfaces, which were verified by experimental synthesis and activity evaluation. The best nanocube and nano-octahedra catalysts were two and three times more active, respectively, than the state-of-the-art commercial pure-platinum catalyst on the basis of activity per gram of platinum, which demonstrates their improved economic viability in consumer markets. In particular, the activity of the octahedra exceeded the 2017 Department of Energy target per mass of platinum used.

In a follow-up work, our team extended this approach to depositing platinum atoms atop palladium icosahedral seeds. This particular geometry allowed further enhancement of the ORR activity by creating a unique platinum overlayer structure. Calculations again elucidated that the enhanced activity arose from compression of platinum in this “wavy” overlayer structure. These catalysts were found to be even more active than the cubes and octahedra, demonstrating performance nearly four times that of the commercial platinum standard. Tests of their durability in the electrochemical environment showed that all three of these particle geometries were much more stable than commercial platinum.

Although these catalysts had greatly improved efficiencies (activity per platinum mass), these core-shell structures still contained a substantial amount of palladium, which has a non-negligible cost of about half that of platinum. We therefore worked to design and synthesize a new class of catalysts, in which the palladium was selectively removed from the core-shell structure (Figure 2). This yielded hollow platinum “shells” only a few atomic layers thick (roughly one nanometer in total thickness). Remarkably, these shells maintained their cubic or octahedral structures, according to the geometries of the initial seeds. This property allowed these structures to maintain their high activities for the ORR; in fact, the ORR activity was further improved relative to the core-shell structures, which our calculations indicated was due to additional compression of the platinum atoms in this novel hollow shell structure. In the end, these were nearly five times as active as the commercial platinum catalyst on the basis of platinum mass; the economics of these catalysts are further improved by the removal of the palladium cores.

Moreover, our calculations identified unique mechanisms by which these ultrathin hollow structures likely form. We predict that platinum and palladium intermix during the core-shell formation process, and that this creates channels of palladium in the platinum shell that ultimately allow the palladium cores to be removed. Further, our calculations indicate that if the shells are too thin, too many channels are formed and the shells lack mechanical stability. If the shells are too thick, then not enough channels can form and palladium cannot be etched from the core. These findings were corroborated by the experimental observations that a critical thickness of the platinum shell is needed to successfully etch hollow structures.

In the future, we foresee the design and synthesis of many new materials, extending into the broader field of catalysis, designed using the principles established in these studies. In addition to predicting new catalysts for improving surface reactivity of chemical reactions, calculations are able to predict the intermixing processes of many transition metal alloys that could be used in a wide range of catalytic reactions, which inform the synthesis community about what methods might be most practical and relevant for obtaining hollow and other novel structures. In turn, the synthesis community continues to develop exciting new protocols for novel catalysts using the design principles set forth by first-principles computations. These integrated cycles of calculation, synthesis, and experimental activity evaluation continue to improve the catalyst design process for the development of new catalysts for sustainable energy and fuels/chemicals production.

Luke Roling is a PhD student in the group of Professor Manos Mavrikakis in the Department of Chemical & Biological Engineering at the University of Wisconsin-Madison.

“The Nexus Blog” is a sister publication of “The Nexus” newsletter. To sign up for the newsletter, please email gci@acs.org, or if you have an ACS ID, login to your email preferences and select “The Nexus” to subscribe.

To read other posts, go to Green Chemistry: The Nexus Blog home.