No abstract

We investigate methane activation over a range of metal-oxide surfaces. Density functional theory calculations are used to correlate the C−H bond activation energy to the surface reducibility (oxygen vacancy formation energy, work function). The correlation includes several reducible and nonreducible metal-oxides, doped CeO 2 , doped TiO 2 , ZnO, and doped MgO, and also holds for various oxidation states of TbO x , different surface facets of TiO 2 , and variation of Hubbard U parameter for CeO 2 . We find a linear correlation between the C−H activation reaction energy, •CH 3 adsorption energy, and the oxygen vacancy formation energy of pure/doped metal-oxides, making surface reducibility a descriptor for predicting catalyst activity and selectivity against further oxidation of the •CH 3 radical.

We examined Ce–Mn mixed oxides as high-temperature desulfurization materials, exploring various Mn/Ce ratios and the effects of admixing other rare earth oxides. The sulfur capacities at temperatures from 900 to 1025 K with simple air regeneration were measured for repeat cycles until a stable, reversible capacity was obtained. The measured sulfur capacities with a realistic model syngas containing H2S, H2, N2, CO, H2O, and CO2 were compared to thermodynamically possible maximum sulfur capacities. The oxidized and sulfided (reduced) sorbents were characterized by X-ray diffraction (XRD), X-ray absorption near-edge spectroscopy (XANES), X-ray absorption fine structure (XAFS), temperature-programmed reduction (TPR), and Brunauer–Emmett–Teller (BET) surface area. Density functional theory calculations are used to aid in interpreting characterization data and in explaining the enhanced S adsorption capacities. There is a large synergistic effect on sulfur adsorption and reaction resulting from the intimate admixing of Mn with CeO2 and CeO2/La2O3 rare earth oxides. However, while these materials are stable at temperatures near 900 K, even using air regeneration, the observed stable sulfur capacities fall far short of predictions based on thermodynamic equilibrium. The differences are attributed to (a) inhibition by CO2 and H2O; (b) formation of some irreversible sulfates upon air regeneration; (c) inability of sulfur to diffuse into larger, sintered crystals of the mixed oxides; (d) gradual dissolution of Mn in an underlying support such as Al2O3 (when present).

MnO(x)-CeO(x) mixed oxide systems exhibit interesting sulfur adsorption capacities and catalytic activity. We examined the electronic structure of Mn-doped fluorite CeO2 bulk solid and surface using density functional theory (DFT) with the Hubbard U term or the Heyd-Scuseria-Ernzerhof (HSE06) hybrid functional. We specifically evaluate the reducibility and formation energies of Mn-doped CeO2 surfaces. The use of a U value on the d-states of Mn is examined, and a value of 4 eV is chosen based upon results from DFT+U calculations on bulk MnO(x),1 XANES characterization of oxidation states in calcined and reduced Mn-doped CeO2, and comparison with HSE06 hybrid functional results. Electronic structure impacts of the U inclusion are discussed. The concentration and orientation of Mn atoms doped into the surface of CeO2 have a great influence on the reducibility of the surface. Based upon formation energies, Mn will not favor doping into the surface of CeO2 in a fully oxidized system (Mn(4+)). Under reducing environments, Mn will dope into the surface with oxygen vacancies present (Mn(3+) and Mn(2+)). The first oxygen vacancy is not likely catalytically important in fluorite MnO(x)-CeO(x) systems as formation of the fully oxidized surface is not stable. A greater degree of reduction would occur during a catalyzed redox reaction.

For CeO 2 or M-doped CeO 2 catalysts, reliable energetics associated with surface reactivity requires accurate representation of oxidized and reduced metal states. Density functional theory (DFT) is used extensively for metals and metal oxides; however, for strongly correlated electron materials, conventional DFT fails to predict both qualitative and quantitative properties. This is the result of a localized electron self-interaction error that is inherit to DFT. DFT1U has shown promise in correcting energetic errors due to the self-interaction error, however, its transferability across processes relevant to surface catalysis remains unclear. Hybrid functionals, such as HSE06, can also be used to correct this self-interaction error. These hybrid functionals are computationally intensive, and especially demanding for periodic surface slab models. This perspective details the challenges in representing the energetics of M-doped ceria catalyzed processes and examines using DFT extensions to model the localized electronic properties.