Site-specific binding maps observed during methanol-to-olefin conversion over ZSM-5 catalysts

The desorption of methanol and dimethyl ether (DME) over ZSM-5 catalysts (Si/Al = 25, 36, and 135) has been investigated by combining temperature-programmed desorption (TPD) studies conducted in a temporal analysis of products (TAP) reactor with a multisite microkinetic modeling approach (fixed-site model, dynamic-site model) based on the mean-field approximation. Six-site ensembles were deconvoluted in line with recent temperature-programmed surface reaction (TPSR) studies and were compared to three-site models using the Redhead method and the microkinetic approach (in the absence of dispersion). These six-site ensemble desorption models give dispersion values with a range of 10–9–10–10 m2 s–1. Fixed-site models and dynamic-site models give similar desorption behavior at full coverage. With all models, methanol has lower desorption energies, requires higher site densities, and has higher desorption rates than does DME. The ratio of binding sites during TPD to binding sites during TPSR for methanol and DME is 1:2 and 2:1, respectively, using a 5 vol % educt. Site-specific scaling relations, site-specific volcano plots, and site-specific binding maps are formed due to opposing site competition and site cooperation and variation in binding energies in accordance with the Sabatier principle. Multiple volcanos on one site ensemble are evinced during DME desorption. Uneven distribution of molecules on sites are predicted. For the first time, using these six-site ensemble models, each site ensemble allows for dissociative adsorption requiring 2 T-sites, thereby connecting macroscopic desorption kinetics with the molecular behavior of 12 crystallographically distinct T-sites, in one configuration, over ZSM-5 catalysts.