Alloys of Pt and Rare Earths for the Oxygen Electroreduction Reaction

This thesis presents the development and characterization of a new class of Pt alloys for catalyzing the Oxygen Reduction Reaction (ORR), in perspective of a future substitution of traditional Pt-based catalysts at the cathode of Polymer Electrolyte Membrane Fuel Cells (PEMFCs). Focused on spectroscopic techniques such as Angle-Resolved X-Ray Photoelectron Spectroscopy (AR-XPS), Ambient Pressure X-Ray Photoelectron Spectroscopy (AP-XPS) and Ion Scattering Spectroscopy (ISS), it takes part in a broader context of studies pursuing the combination of these physical techniques with electrocatalysis.

A number of bimetallic alloys based on Pt and a rare earth, like the Pt-Y system or more recently proposed Pt-lanthanide phases, have been tested and characterized. Polycrystalline Pt5La and Pt5Ce exhibited more than a factor of 3 enhancement in specific activity relative to state-of-the-art polycrystalline Pt.

They maintain at least 90% of this activity after accelerated stability tests (10 000 cycles between 0.6 and 1.0 V vs. the Reversible Hydrogen Electrode (RHE) in 0.1 M HClO4 electrolyte). A combination of AR-XPS and ISS measurements allowed to elucidate the active surface phase and structure of these materials, consisting of a ˜1 nm thick pure Pt overlayer on top of the bulk alloy, the stable overlayer providing kinetic stability against further dissolution of the lanthanides.

We hypothesize that this high stability is related to the very negative heat of formation of their intermetallic phases, that would prevent La and Ce diffusion to the surface. For this structure only strain effects can explain the activity enhancement. Other alloys of the same class (e.g. Pt5Gd and Pt5Tb) exhibit even higher specific activities, up to 6 times the one of polycrystalline Pt in the case of Pt5Tb, a record activity among polycrystalline alloys.

On the basis of their similar crystal structures, the ORR activity of this class of alloys is correlated to the lattice parameter of the bulk, which is expected to define the Pt-Pt distance in the overlayer. The compression of this Pt-Pt distance in the overlayer originates a volcano-shape trend in activity.

However, the most active alloys experience higher activity losses during stability tests, suggesting that high levels of compression might not favour the long-term stability of the Pt overlayers. This hypothesis is supported by Density Functional Theory (DFT) calculations and by AR-XPS. A model for the quantitative estimate of the Pt overlayer thickness from AR-XPS measurements indicates a correlation between the thickening of the Pt overlayers and the activity losses, supporting the concept that more compressed overlayers have lower physical stability.

The application of these materials in a fuel cell requires the fabrication in nanoparticulate form. Through the combination of a gas aggregation technique and a time-of-flight mass spectrometer size-selected Pt-Y nanoparticles are produced. With a mass activity of 3.05 A mg-1Pt at 0.9 V vs. RHE, 9 nm Pt-Y nanoparticles are among the most active ORR catalysts ever reported, although they lose 37 % of this activity after stability test.

Similar to the case of polycrystals, after immersion in the acidic electrolyte and testing the active phase consists of a Pt shell surrounding an alloyed core. Also in this case the compressed Pt-Pt distance explains the ORR activity enhancement of these catalysts. The deposition of these 9 nm Pt-Y nanoparticles on the cathode side of a Membrane Electrode Assembly (MEA), part of a specifically prepared fuel cell, allows AP-XPS measurements under operation conditions.

As a consequence of potential cycling, Y oxidizes due to the dealloying process which is observed in-situ. The adsorbed species can be also probed and correlated to the electrochemical potential. Near the open circuit potential (OCP) conditions the oxygenated species consist, to a good extent, of non-hydrated OH, similar to the case of pure Pt nanoparticles.