Contrasting Catalytic Functions of Metal Vanadates and Their Oxide Composite Analogues for NH3-Assisted, Selective NO_x Transformation

摘要

V2O5 fuses with transition metals to create dozens of different metal vanadates, whose acidic/redox traits can be diverse yet optimized for selective catalytic NO_x reduction (SCR) by changing the metals used or their metal:vanadium stoichiom-etry. However, no metal vanadate has been compared with its metal oxide composite analogue as an active phase for SCR, albeit a vanadate occasionally outperforms an oxide composite simulating a commercial catalyst (V2O5-WO3). Herein, Cu3V2O8 and CuO-VO2/V2O5 were rationally selected as model phases of metal vanadates and oxide composites and isolated using pH regulation of their synthetic mixture to ≤～5 (pHl/pH5) and ～11 (pH11), respectively. The pH1/pH5/pH11 samples were comparable with regard to morphological, textural, and compositional traits but not for crystallographic features. This thus provided the impetus to simulate the pH1/pHS/pH11 surfaces under a SO2-containing feed-gas stream, by which SO_A~(2-)/HSO_A~- functionalities (A = 3-4) were anchored on their (defective) Lewis acidic metals and/or labile oxygens (O_α). This could result in the formation of pH1-S/pH5-S/pH11-S, whose major surface species were Bronsted acidic bonds (SO_A~(2-)/HSO_A~-) and redox sites (O_α; mobile oxygen (O_M); oxygen vacancy (O_V)). pH1-S/pH5-S/pH11-S were similar in terms of NH3 binding energies and energy barriers in SCR yet escalated collision frequencies among the surface species involved in the sequence of pH11-S < pH5-S < pH1-S (via kinetic assessments), as was the case with the numbers of SO_A~(2-)/HSO_A~- functionalities of the catalysts (via temperature-resolved Raman spectroscopy). These were coupled to elevate the efficiency of acidic cycling on the order of pH11-S < pH5-S < pH1-S. Meanwhile, the amounts of O_α and O_V (or O_M) innate to pH1-S/pH5-S were smaller than and comparable to those of pH11-S, respectively. Nonetheless, pH1-S/pH5-S provided greater O_M mobility than pH11-S, thereby proceeding better with redox cycling than pH11-S (via ~(18)O-labeling O2-on/off runs). Furthermore, pH1-S/pH5-S outperformed pH11-S in SCR under diffusion-limited domains, while enhancing the resistance to H2O, ammonium (bi)sulfate poisons, or hydrothermal aging over pH11-S by diversifying the selective N2 production pathway other than SCR.