Competing adsorption of H and CO on Pd-alloy surfaces: Mechanistic insight into the mitigating effect of Cu on CO poisoning

Abstract

Multi-component alloys offer broad tunability for addressing challenges in materials science, but their vast configurational space makes their surface chemistry highly sensitive to operating conditions, for example through adsorption and segregation. Here, we study Pd-Au-Cu alloy surfaces in H$_2$ and CO environments motivated by their use in H technologies, in particular plasmonic H$_2$ sensing, where alloying can mitigate limitations intrinsic to Pd such as hysteresis and CO poisoning. Modeling multicomponent surfaces with multiple adsorbate species under realistic conditions is challenging. To this end, we establish an accurate and efficient framework that combines machine-learned interatomic potentials trained on density functional theory data to generate training data for cluster expansions with effectively no limitations on training set size. By constructing continuous surface phase diagrams for H-CO coadsorption we find that coadsorption under operating conditions is governed primarily by the H coverage during annealing. Au-rich surfaces, formed under H-poor conditions, suppress both CO and H adsorption, while H-rich conditions yield Pd-rich surfaces that maintain higher H coverages compared to Pd at relevant CO partial pressures, indicating improved CO poisoning resistance. This effect is insensitive to relative amounts of Au and Cu, despite experimental evidence of the mitigating effect of specifically Cu on CO poisoning. Kinetic barriers for dilute alloy surfaces indicate that absorption pathways near Au are highly unfavorable, while Cu leave the energetics unchanged compared to pure Pd. This finding suggests that Cu in the surface region provides viable pathways to shuttle H into the material when Pd-dominated paths are blocked by CO.

Figures (11)

• Figure 1: Model development. (a) Schematic illustration of the computational framework. (b) Model errors for the and the . Note that the data has been systematically shifted along the y-axis to allow for clearer visualization of the different models.
• Figure 2: Schematic illustration of the surface slabs and their respective adsorption sites. Side views of the surface slabs (to the left) with the surface layers and bulk region indicated. Top views of the surfaces (to the right) with blue atoms for the top layer, gray atoms for the subsequent layers. The smaller atoms show the adsorption sites labeled by their first letter. The arrows between sites indicate their three-dimensional assuming a 4.0 lattice parameter.
• Figure 3: Model performance for surface properties. (a) Segregation energies from , , MACE, and for dilute Pd alloys in vacuum, H2 and CO (100 coverage). Negative (positive) values indicate that the minority species prefers (avoids) residing in the surface region. Note that the curves are shifted along the y-axis for better visualization. (b) Adsorption energies for H and CO on Pd (25 coverage) for all models. Additional sites and adsorption energies on Cu and Au can be found in \ref{['sfig:adsorption-energies-all']}.
• Figure 4: construction for Pd {111}. Traditionally, are constructed by calculating the free energy as a function of the chemical potential for specific coverages (a) and translating the convex hull to a discrete surface diagram (b). sampling of allows for efficient sampling of surface coverage(s) as a function of chemical potential(s) which results in continuous surface diagrams (c).
• Figure 5: Conversion between chemical potential and pressure for Pd. (a) H coverage vs. chemical potential obtained from simulations. (b) Experimental records of H coverage vs. partial pressure from Mitsui 03 MitRosFom03, Engel 79 EngKui79, and additional sources (see \ref{['sfig:pressure-conv']}) WilMatFuk01. (c) Fit of the reference chemical potential based on the data in (a) and (b) as well as the calculated reference chemical potential based on tabulated thermodynamic data (JANAF nist-janaf) or computational models (Rahm 21 RahLofFra21). Once the reference chemical potential is known, the data (a) can be converted from chemical potential to partial pressure, as shown in (b) for the different references. (d--f) show the same process for CO with experimental data for Pd:CO from Unterhalt 02 UntRupFre02, Kuhn 92 KuhSzaGoo92, Kaichev 03 KaiProBuk03, and additional sources (see \ref{['sfig:pressure-conv']}) HeMemGri88HeNor88BehChrErt80. Note that the legends are shared across each row.