No abstract

Some catalytic oxide supports are more equal than others, with numerous variable properties ranging from crystal symmetry to surface chemistry and electronic structure. As a consequence, it is often very difficult to determine which of these act as the driver of performance changes observed in catalysis. In this work, we hold many of these variable properties constant with structurally similar LnScO3 (Ln = La, Sm, and Nd) nanoparticle supports with cuboidal shapes and a common Sc-rich surface termination. Using CO oxidation over supported Au nanoparticles as a probe reaction, we observe higher activation energy and a slower rate using NdScO3 as the support material. This change is found to correlate to the strength of CO2 binding to the support surface, identified by temperature-programmed desorption measurements. The change is due to differences in the 4f electrons of the lanthanide cations, the cations’ Lewis acidity, and the inductive effect they impose.

Bulk and supported MoS x materials have gained interest as alternative catalysts for light alkane dehydrogenation, but there is little kinetic data published with which rigorous comparisons can be made to other catalysts. Here, rates, selectivities, and activation barriers are collected under conditions of differential conversion for the dehydrogenation of isobutane over various morphologies of molybdenum sulfide. We find that a "rag-like" MoS 2 composed of small, disordered crystallites exhibits higher catalytic rates than layered, highly crystalline MoS 2 (52 vs 2.7 μmol ks −1 g cat −1 at 360 °C). This is in part not only due to increased surface area but also due to intrinsically more active edge and defect sites exposed by the rag-like structure, as shown by a decrease in apparent activation energy from 61 to 43 kJ mol −1 . Supporting MoS x on metal oxides or metal organic frameworks gives small MoS x clusters that have up to 7-fold higher rates per mass of MoS 2 than even the rag-like MoS 2 . Rates are support-dependent, with the highest rates per mass of MoS 2 observed over MoS x /TiO 2 . Pt/Al 2 O 3 has a ∼50-fold higher rate than the best MoS 2 catalysts (2700 μmol ks −1 g cat −1 at 360 °C), but it has an apparent activation energy of 41 kJ mol −1 , similar to that of the rag-like MoS 2 . Therefore, the rates over MoS 2 appear to be limited by a small number of active sites on the surface, rather than intrinsically poor activity. Given the data provided in this manuscript and the enormous phase space available to metal sulfides, these materials warrant further investigation as alternative light alkane dehydrogenation catalysts, especially for use under conditions that would deactivate a precious metal catalyst.

No abstract

No abstract