Overoxidation is the principal barrier against commercializing propane oxidative dehydrogenation (PODH) catalysts for propylene production. The current approach to reducing overoxidation, i.e., coating the non-selective support surface with a monolayer of active phase, can itself increase the probability of overoxidation of the produced propylene due to polymerization of active phase species. Incorporating the "confinement agents" onto the metal oxide support might be considered as an alternative solution to prevent hydrocarbons from reaching the support and overoxidizing. Herein, the UiO-66 metal-organic framework, which contains numerous organic ligands connected to the zirconia nodes, was used as support for the vanadia active phase to highlight the role of support's confinement effects on the overall catalytic performance toward the PODH. The UiO-66 supported vanadia catalysts with various vanadium loadings were fabricated via an ultrasonic-assisted wet impregnation procedure. The catalytic function is related to the underlying chemical processes at catalyst surfaces using physicochemical characterization techniques, PODH performance measurements, and machine learning tools. The results showed that the catalyst with a relatively low vanadia density of 2.7 nm-2, equivalent to less than half of the entire support surface coverage, could achieve propylene productivity of 4.43 [Formula: see text] , propane conversion of 17.1%, and propylene selectivity of 49.7% at 350 °C.
Keywords: Machine learning; Metal–organic framework; Oxidative dehydrogenation; Propane; Propylene; UiO-66; Vanadium oxide.
Copyright © 2022 Elsevier Inc. All rights reserved.