Understanding the Poisoning, Aging and Degradation of Low-temperature Fuel Cell Electrocatalysts Using in situ X-ray Absorption Spectroscopy Open Access
Downloadable ContentDownload PDF View PDF in Browser Report an accessibility issue with this item
In situ x-ray absorption spectroscopy (XAS) was employed in this dissertation to probe the poisoning and degradation of platinum (and Pt-based) electrocatalysts under realistic operating conditions. XAS is a high-energy spectroscopic technique able to provide both structural and electronic information, and therefore is uniquely suited to study electrocatalysts under operating conditions. Conventional EXAFS analysis was combined with the novel delta-mu XANES method to provide both nanoparticle morphology and adsorbate coverages on commercially available fuel cell catalysts. The spectroscopic data were complemented with data from electrochemical techniques such as cyclic voltammetry (CV) and chronoamperometry (CA), along with rotating disk electrode (RDE) and copper underpotential deposition (Cu upd) experiments. The loss of catalytic activity in a fuel cell with age occurs through two chief processes: poisoning of active surface sites and loss of surface sites through particle morphological changes, coalescence and aggregation. All of these processes were investigated using spectroscopic and electrochemical techniques. The poisoning of Pt/C electrocatalysts by chloride and ruthenium ions was studied using in situ XAS. RDE experiments show unequivocally that adsorbed chloride drastically hinders the Pt reactivity by blocking active surface sites and by increasing the overpotential for the oxygen reduction reaction (ORR) by approximately 85 mV for every 10-fold increase in chloride concentration. Through the use of the delta-mu XANES method, we were able to provide direct spectroscopic evidence for site-specific adsorption of Cl ions on the 3-fold sites of the (111) planes of Pt nanoparticles, although some of the adsorbed chloride are forced into bridged or atop sites by strong lateral interactions at high Cl coverage. It has been established in the literature that ruthenium ions are released into the electrolyte as a result of degradation of PtRu anode catalysts. The electrochemistry, electron-spin resonance (ESR) and XAS results reported in this thesis collectively confirm earlier findings that these species travel through the polymer membrane and deposit onto Pt/C cathodes, decreasing the ORR activity. ESR results show that the Ru ions deposited in the membrane alter the hydration levels and transport properties of the membrane. The deposition of Ru on the Pt cathode was found to be most severe at open circuit potential (ca. 0.95 V vs. RHE), when the surface is partially covered with O anions, which may induce a Coulombic attraction for the Ru cations. Comparisons between the experimental delta-mu XANES results and full multiple scattering calculations using the FEFF 8.0 code on Pt6 model clusters suggest that the Ru species adsorb primarily in 3-fold sites on the Pt surface. Semi-quantitative estimates of the Ru coverage on the Pt/C catalysts, the first such estimate using XAS, are shown to be in good agreement with other studies in the literature. An in situ XAS study at both the Pt L3 and Ru K edges on the stability of two commercial PtRu catalysts aged through voltammetric cycling, along with chronoamperometry results, reveals that the initial morphology of the PtRu nanoparticles plays a major role in the catalysts long term stability. A delta-mu XANES analysis was carried out to follow the site number changes with aging, while EXAFS analysis provided structural information on the changing composition and morphology of the catalysts. It was found that the samples with larger, more oxidized Ru islands on the nanoparticle surface are less susceptible to Ru dissolution than those with smaller, more metallic Ru islands. Further, as expected, the smaller Ru islands grew faster than their more oxidized counterparts. These findings and other insights provide an increased understanding of the observed changes in the methanol oxidation CVs with aging. Polyvinyl Pyrrolidone (PVP) is a widely used organic capping agent that is also used to prevent coalescence and aggregation in nanoparticles, effectively slowing the aging process that commonly occurs during catalysis. While it is generally accepted that a small amount of PVP (ca. 5-10 wt. %) remains closely associated with the synthesized nanoparticles to retain their shape, it has been generally assumed that the PVP itself does not alter the catalytic activity of the Pt. We report that PVP-capped Pt/C nanoparticles display a remarkable enhancement of their methanol oxidation activity over plain Pt/C nanoparticles with the exact same size and nanostructure, corroborating a recent study on Pt black catalysts. Thus the PVP capping agent not only stabilizes the nanoparticles against aging, but also plays a role, presumably through a ligand effect, in enhancing the Pt catalytic activity for certain reactions. An in situ XAS study aimed at directly probing the PVP-Pt interaction reveals that this interaction is potential-dependent: a more neutral PVP-Pt interaction (PVP-N) exists at lower potentials (V < 0.60 V vs. RHE) and changes to a stronger interaction at higher potentials, involving charge-transfer (PVP-CT) from the PVP to Pt. Theoretical FEFF 8.0 calculations modeling the PVP-CT/Pt suggest that the PVP-CT bonds to platinum in atop sites, while the PVP-N appears to be either more mobile or not in registry on the surface. Further, CV data suggests that the PVP-N preferentially blocks H adsorption at sites on the (100) faces.
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.