Understanding the interaction between reactant molecules and “single atom” active sites is important for comprehending the evolution of single-atom catalysts in reactive atmospheres. Here, we study how Fe₃O₄(001)-supported¹ Rh₁ monomers and Rh₂ dimer species interact with CO using density functional theory (DFT), combined with temperature-programmed desorption, x-ray photoelectron spectroscopy, and in-situ scanning tunneling microscopy techniques. Our results show that stable Rh₁(CO)₁ monocarbonyls are the exclusive product of CO adsorption at both 2-fold and 5-fold coordinated Rh₁ sites, but the different coordination environment leads to different adsorption energies. The DFT calculations reveal that the Rh₁(CO)₁ formed at the 5-fold coordinated Rh₁ site adopts an octahedral structure, while the Rh₁(CO)₁ formed at the 2-fold coordinated Rh₁ site forms an additional bond to a subsurface oxygen atom of the support, leading to a pseudo-square planar structure. The direct addition of a second CO molecule to Rh₁(CO₁) at the 2-fold coordinated Rh₁ site to form a Rh₁(CO)₂ gem-dicarbonyl is energetically favorable according to DFT; however, this process is not observed in experiments under UHV conditions. Instead, we observe the formation of limited Rh₁(CO)₂ exclusively via the CO-induced breakup of Rh₂ dimers, in agreement with DFT results, which suggest an unstable Rh₂(CO)₃ intermediate.

1. Bliem, R. et al. Subsurface cation vacancy stabilization of the magnetite (001) surface. Science 346, 1215–1218 (2014).

In the past, single atom catalysts (SACs) could not be clearly visualized and characterized due to the limitations associated with instrumental resolution. Today this is a new frontier in heterogeneous catalysis due to the high activity and selectivity of SACs for various catalytic reactions. This has opened various questions for theory. One is where are the atoms and what is the stability of SACs in working conditions. In order to address these questions, we will discuss the nature of isolated metal species deposited on oxide surfaces (TiO₂ and ZrO₂ in particular). These systems have been characterized experimentally using high-resolution scanning transmission electron microscopy (STEM), Fourier transform infrared spectroscopy (FTIR), and temperature programmed desorption (TPD) spectra of adsorbed CO probe molecules. Combining these data with extensive Density Functional Theory (DFT) calculations one can provide an unambiguous identification of the stable single-atom species present on these supports and of their dynamic behavior.

The other question that can be addressed by theory is the prediction of the behavior of SACs in electrocatalytic processes such as the oxygen reduction (ORR), the oxygen evolution (OER) and the hydrogen evolution (HER) reactions. In this context we assist to a rapidly growing number of DFT studies and of proposals of universal descriptors that should provide a guide to the experimentalist for the synthesis of new catalysts, in particular related to graphene-based SACs. We will critically analyze some of the current problems connected with these DFT predictions: accuracy of the calculations, neglect of important contributions in the models used, physical meaning of the proposed descriptors, inaccurate data sets used to train machine learning algorithms, not to mention some severe problems of reproducibility. It follows that the “rational design” of a catalyst based on some of the proposed universal descriptors or on the DFT screening of large number of structures should be considered with some caution.

The hydroformylation of alkenes has emerged as one of the most interesting applications of “single-atom” catalysis. Nevertheless, there have been relatively few fundamental studies into how the reactants (CO, alkene, and H₂) bind at the active site. In this talk I will show STM, XPS, TPD and DFT results to illustrate how C₂H₄ and H₂ interact with a Rh₁/Fe₃O₄(001) model catalyst. Ethylene binds strongly at the Rh₁ sites, but there is very little evidence for the formation of di-ethylene species under UHV conditions. H₂ adsorbs as a dihydride at the Rh₁ sites, and desorbs close to room temperature in TPD experiments without spilling over onto the oxide support. Evidence for the co-adsorption of the different reactants will be discussed in the context of the hydroformylation reaction.

Producing fuels and chemicals using renewable electricity holds the promise to enable a truly sustainable circular economy based on sustainably produced carriers of electrical energy and sustainably produced chemicals. To date, the vast majority of electrocatalytic reactions are limited to the transformation of small inorganic molecules such as CO₂, H₂O, N₂, as well as the oxidation and reduction of alcohols. However, comprehensive electrification of the chemical industry will require electrocatalytic reactions that can promote the transformations of C(sp³)-H and C(sp³)-C(sp³) bonds, which are central to today’s industry.

In this presentation, I will show how fundamental understanding of the interfacial processes occurring in electrocatalytic reactions can be exploited to expand the reaction scope of electrocatalysis to the transformation of complex substrates involving the controlled activation of C-H and C-C bonds. In a first step, I will show how this approach allows us to transform ethanol to ethylene oxide, an important plastic precursor. Subsequently, I will discuss methods to electrocatalytically transform inert alkanes such as methane and ethane at room temperature.

A class of catalysts called single-atom alloys allow for the combination of a more reactive, more expensive dopant metal dispersed within a less active, more selective, and cheaper base metal. These catalysts have been well characterized and studied using techniques such as temperature programmed desorption (TPD), scanning tunneling microscopy (STM), and reflection absorption infrared spectroscopy (RAIRS). However, the detailed, molecular-level bond activation energetics and kinetics have not yet been experimentally interrogated for high-barrier reactions on these catalysts—a region where more efficient catalysts are most sorely needed.

We will present recent results that first characterize the dissociation and spillover of H resulting from both atomic and molecular H2 adsorption on base Cu(100) and RhCu(100) single atom alloy, and then describe results from energy resolved molecular beam studies of CH4 dissociation that quantify reaction probability as a function of energy distribution among reactant and surface degrees of freedom. We expect these studies will reveal new insights into the molecular mechanism for an important class of heterogeneously catalyzed reactions and provide new benchmarks for computational studies of single atom catalysts.