Earth-abundant metal oxides expose Lewis acid-base (LAB) site pairs that heterolytically activate H–H and C–H bonds and exhibit dehydrogenation and hydrogenation gravimetric rates comparable to dispersed metal nanoparticles. The most reactive sites for kinetically-relevant C–H activation exist as minority species (<0.6 nm^-2) and strongly bind H₂O, rendering experimental determinations of their coordinative properties impractical which challenges the design of highly-reactive catalysts. Here, Density functional theory (DFT) estimates of the energies of bound intermediates and transition states (TS) identify geometric arrangements that best stabilize TS structures for H–H, O–H, and C–H activation by probing different Zr–O pairs at low-index planes in monoclinic and tetragonal ZrO₂ surfaces. These models reveal that TS and product (PS) energies for H₂, H₂O, and C_nH_2n+2 dissociations are not related by a single-value function, precluding the use of scaling relations (from PS energies) that avoid computationally-intensive TS energy calculations. Such relations are frequently applied to homolytic bond activation at metal surfaces, but the geometric features of binding sites in LAB pairs effect the energies of bound reactants, products, and TS inconsistently due to concerted interactions of adsorbate species with site ensembles and through-space electrostatic stabilization. Significant changes in the geometry of adsorbates and the LAB site pairs arise between the TS and PS in heterolytic dissociation. Further, TS energies are influenced by electrostatic stabilization that fully develops in the PS yet is not detected by TS bond distances of the cleaving moieties. Ionic stabilization varies significantly among adsorbates due to binding geometry constraints imposed by the unique orbital symmetry of each adsorbate. As a result, molecule and site properties alone are insufficient as independent descriptors of reactivity. These findings highlight the necessity of explicit TS assessments and identification of the most competent sites requires models that accurately reflect the active sites in physical systems.

Janik Hense

Appearances