463595 Methanol Oxidation on Pristine and Doped MoO3 (010): A DFT and Microkinetic Analysis

Wednesday, November 16, 2016
Grand Ballroom B (Hilton San Francisco Union Square)
Tej S. Choksi and Jeffrey P. Greeley, School of Chemical Engineering, Purdue University, West Lafayette, IN

Methanol oxidation is employed as a probe reaction to rigorously evaluate catalytic properties of molybdenum trioxide (MoO3), a reducible oxide that exhibits a fascinating interplay of catalytic chemistry and structural transformations. As a first step towards this goal, the catalytic cycle is studied on the (010) basal planes, which exist during the early stages of methanol oxidation. The reaction mechanism is investigated with electronic structure calculations using the BEEF-vdW and HSE06 functionals along with mean field microkinetic modeling. Considered pathways include vacancy formation and oxidation, monomolecular dehydrogenation steps on reduced and non-reduced surfaces, bimolecular reactions between dehydrogenated intermediates, and precursor steps that form hydrogen molybdenum bronze phase (HyMoO3-x). Methanol dissociation begins with C-H or O-H scission, leading to two distinct pathways, with the O-H route found to be kinetically and thermodynamically more favorable. Dehydrogenation of CH2O* to CHO* is kinetically slower as compared with desorption (by a factor of 1.4x105), leading to complete selectivity towards CH2O. The most favorable pathway has three kinetically significant elementary steps: (1) C – H scission of CH3O*, (2) recombination of dissociated OH* to form H2O*, and (3) dissociative adsorption of Owhich oxidizes the surface. (1) and (2) promote methanol oxidation, while (3), oxidation of the reduced surface is rate inhibiting. Reaction orders with respect to gas phase methanol and oxygen are 1.0 and -0.5 respectively.

The surface state of a reducible oxide can have a direct impact on Mars van Krevelen-type catalysis. Thus, the microkinetic analysis is applied towards understanding the relationship between catalysis and surface reducibility by varying the oxygen partial pressure. As a consequence of too few vacancies limiting CH2O and vacancy formation at atmospheric pressure, and too low O2* coverage limiting surface oxidation in Ultra High Vacuum (UHV), the reaction rate follows a volcano dependency across a range of oxygen pressure, with the optimum rate located where surface oxidation is neither limiting nor inhibiting the overall rate. Thus, at the volcano maximum, the reaction order with respect to gas phase oxygen is close to zero. Finally, a Constable – Cremer relationship (compensation effect) between the apparent prefactor and apparent activation energy is observed as the mechanism changes around the top of the volcano.

A simplified analytical form of the rate expression, which is valid for a wide range of oxygen pressure, is proposed in this contribution. The kinetic and thermodynamic constants in this expression can be tuned through geometric (strain, coordination environment) and electronic modifications (substitutional doping) of the (010) surface of MoO­3, suggesting strategies to enhance catalytic activity of the basal planes of MoO3. The DFT and microkinetic modeling approach demonstrated here is generally suitable for analysis of catalytic mechanisms on reducible oxide surfaces.

Extended Abstract: File Not Uploaded