AVS 66 Session HC+2D+SS-WeM: Exotic Nanostructured Surfaces for Heterogeneously-Catalyzed Reactions

Selective conversion of alkanes to higher value species using heterogeneous catalysts is of great interest with the increasing availability of light alkanes from shale fracking. We have used a combination of temperature programmed reaction spectroscopy (TPRS) and density functional theory (DFT) to demonstrate that the stoichiometric terminated IrO₂(110) surface can activate methane and ethane below room temperatures, and furthermore, that this surface can be selective towards ethane dehydrogenation to ethylene. For ethane, DFT shows that adsorption and initial C-H bond cleavage to surface bound C₂H₄* is facile and the selectivity step occurs between further C-H bond breaking leading to complete oxidation versus ethylene desorption. The reactivity of this surface is mediated by the presence of undercoordinated Ir (Ir_cus) and adjacent bridge O atoms (O_br). Using the combination of TPRS and DFT we find that pre-hydrogenating the IrO₂(110) surface results in the formation of HO_br sites that increases the selectivity towards ethylene by increasing the barrier to C-H bond cleavage for C₂H₄* and decreasing the desorption energy of C₂H₄*. We will discuss efforts to use DFT and microkinetic modeling to explore doping strategies of both the Ir_cus and O_br sites to promote selectivity towards ethylene formation.

Tuning catalysts for selective hydrogenation reactions is ultimately determined by the nature of the active site for H₂ dissociation and the adsorption of atomic hydrogen on the surface. Several single atom alloys (SAAs) consisting of small amounts of Pd deposited onto surfaces of metals that do not activate H₂ dissociation, such as Cu(111) and Au(111), have been previously studied. In the present study, we characterize Pd/Ag(111), a possible new single atom alloy surface using reflection absorption infrared spectroscopy (RAIRS) of adsorbed CO as a probe. From 0.01 to 0.04 ML Pd/Ag(111), a ν(CO) stretching peak was seen at 2050 cm^-1 corresponding to CO adsorbed on palladium atoms at the on-top site, indicating a single atom alloy surface. By increasing the palladium coverage to approximately 0.05 ML and above, a second ν(CO) stretching peak was seen at 1950 cm^-1 corresponding to CO adsorbed on a palladium bridge site, indicating palladium dimer formation. The surface palladium coverage was determined using temperature programmed desorption (TPD) of CO and Auger electron spectroscopy (AES). By annealing these surfaces to 500 K, the palladium atoms diffuse into the subsurface, and a ν(CO) stretching peak at 2150 cm^-1 (CO adsorbed on silver atoms) is greatly enhanced in intensity due to subsurface palladium. The subsurface to surface palladium ratios on the single atom alloy surfaces were varied from capped Ag/Pd/Ag(111), to a 50:50 ratio, to approximately a 60:40 ratio. The ability of subsurface palladium on the Pd/Ag(111) SAA surfaces to facilitate hydrogen dissociation was explored using H₂ and D₂ TPD.

The hydrogenation of propyne (C₃H₄) to propylene (C₃H₆) using a Pd/Cu(111) single atom alloy (SAA) has been studied using polarization dependent-reflection absorption infrared spectroscopy. This method allows for simultaneous monitoring of reactants and products in the gas-phase and species adsorbed on the surface during the reaction. The results were compared with the hydrogenation of propyne using Pd-free Cu(111) as well as previous studies on Pd/Cu SAA alumina-supported metal catalysts. Propylene production first occurs at 383 K as indicated by the presence of an infrared peak at 912 cm ^-1, which is a uniquely characteristic of gas-phase propylene. The presence of propyne oligomers on the surface is indicated by a dramatic increase in the peak intensity at 2968 cm ^-1 at temperatures above 400 K. The progression of the peaks at 912 and 3322 cm^-1 was used to calculate the rate of production of propylene and the rate of consumption of propyne, respectively. This reaction rate was used to determine a turnover frequency (TOF) for the reaction on the Pd/Cu SAA catalyst.

The field of „single-atom” catalysis (SAC) [1-2] emerged as the ultimate limit of attempts to minimize the amount of prescious metal used in heterogeneous catalysis. Over time, it has become clear that metal adatoms behave differently to supported nanoparticles [3-4], primarily because they form chemical bonds with the support and become charged. In this sense, SAC systems resemble the mononuclear coordination complexes used in homogeneous catalysis, and there is much excitement that SAC could achieve similar levels of selectivity, and even heterogenize problematic reactions currently performed in solution. It is important to note, however, that homogeneous catalysts are designed for purpose based on well-understood structure-function relationships, but the complexity of real SAC systems means that the structure of the active site is difficult to determine, never mind design. In this talk, I will describe how we are using precisely-defined model supports [5] to unravel the fundamentals of SAC. I will show a selection of our latest results in this area, including scanning probe microscopy, x-ray photoelectron spectroscopy (XPS) and temperature programmed desorption (TPD) data to show how the local structure of Ir₁/Fe₃O₄(001) and Rh₁/Fe₃O₄(001) single atom catalysts changes based on preparation and adsorption of reactants, and how the structures obtained can be rationalised by analogy to Ir(I) and Ir(IV) complexes, respectively. If time permits, I will also show that CO oxidation activity in the Pt₁/Fe₃O₄(001) system is promoted by water.

[1] Qiao, B., et al., Single-atom catalysis of CO oxidation using Pt1/FeOx. Nature Chemistry 3 (2011) 634-41.

[2] Liu, J., Catalysis by supported single metal atoms. ACS Catalysis 7 (2016) 34-59.

[3] Gates, B.C., et al., Atomically dispersed supported metal catalysts: perspectives and suggestions for future research. Catalysis Science & Technology 7 (2017) 4259-4275.

[4] Parkinson GS, Catalysis Letters 149 (2019), 1137-1146

[5] Bliem, R., et al., Subsurface cation vacancy stabilization of the magnetite (001) surface. Science, 346 (2014) 1215-8.