Structure and behavior of Mo-zeolite catalysts for non-oxidative methane dehydroaromatization

Abstract

Methane is an abundant component of natural gas, yet a significant portion of the global supply is stranded in remote reserves and ultimately wasted through flaring or venting. A particularly promising route to valorize methane on-site is its direct, non-oxidative conversion to liquid aromatics via methane dehydroaromatization (MDA), a reaction catalyzed at 700 °C and atmospheric pressure by molybdenum-modified MFI-type zeolite catalysts (Mo/MFI). Rapid deactivation of Mo/MFI remains a significant barrier to industrial implementation of the MDA process, and the mechanistic origins of this deactivation are obscured by a lack of consensus surrounding the true nature of MDA active sites and the mechanisms of reaction and deactivation. This dissertation aims to advance understanding of Mo speciation in Mo/MFI and its role in the catalytic activity and deactivation during MDA, all throughout the catalyst's life cycle. The speciation of precatalytic Mo-oxide sites on Mo/MFI was first investigated as a function of catalyst preparation method and composition (Mo loading, zeolitic Brønsted acid site density). Three distinct precatalytic Mo-oxide structures (MoO2OH+ monomer, MoO22+ monomer, Mo2O5+ dimer) were found to coexist on the same catalyst, with their distributions affected by choice of synthesis technique and catalyst composition. This finding reconciles long-standing discrepancies in the literature regarding the precatalytic structure of Mo/MFI catalysts. Kinetic evaluation of catalysts with different starting Mo-oxide populations yields virtually identical catalytic and deactivation behavior in reaction, suggesting that activation of MDA catalysts results in active sites with convergent kinetic properties. Next, H2–temperature programmed reduction (H2–TPR) was revisited as a quantitative probe of Mo location within Mo/MFI, using a series of catalysts spanning a wide range of Mo dispersions to allow for the distinction between internal and external Mo sites across catalysts of varying Si/Al ratio. This distinction is otherwise confounded by the inherent coupling between the Mo-site distribution and BAS density, which complicates kinetic evaluation of distinct site populations. Correlation with reaction data showed that only internal Mo sites contribute to benzene formation above a minimum threshold, regardless of changes in external Mo loading or Si/Al ratio, pointing towards a monofunctional reaction mechanism dictated solely by the internal Mo-carbide sites. This finding establishes a consistent quantitative basis for tracking the active Mo-site population across catalyst compositions. Finally, Mo/MFI deactivation during MDA was investigated on the basis of two primary contributions, outward Mo-site migration and coke deposition, using the internal/external Mo quantification methodology developed in the preceding chapter in parallel with quantitative coke analysis on spent catalysts. Kinetic studies showed that the transition between "fast" and "slow" deactivation regimes occurs at approximately the same time-on-stream across different space velocities. This transition time coincides with saturation of the outward Mo migration process, which appears insensitive to catalytic turnover. Beyond this point, continued activity loss is controlled by coke deposition, which proceeds at a slower but sustained rate and exhibits a clear dependence on space velocity. From these observations, a mechanistically informed, dual-term model for MDA activity of the form: a(t) = (1 − dmig)(1 – dcoke) is proposed for the first time. The model is shown to independently recover deactivation kinetic parameters from activity data alone, in quantitative agreement with those derived from rigorous experimental characterization. This provides a facile yet physically meaningful diagnostic tool for quantifying the separate Mo migration and coking contributions to Mo/MFI deactivation, informing the design of operating conditions and regeneration strategies that target the dominant deactivation mode for a given regime.