(356a) Kinetic Modeling of Nitrous Oxide Decomposition on Fe-Zsm-5 Based on Parameters Obtained from a Comprehensive Dft Study | AIChE

(356a) Kinetic Modeling of Nitrous Oxide Decomposition on Fe-Zsm-5 Based on Parameters Obtained from a Comprehensive Dft Study

Authors 

Heyden, A. - Presenter, University of South Carolina
Bell, A. T. - Presenter, University of California, Berkeley
Keil, F. J. - Presenter, Hamburg University of Technology


Fe-ZSM-5 is an active catalyst for the stoichiometric decomposition of N2O to N2 and O2 and is therefore potentially useful for the abatement of N2O emissions from industrial waste streams, such as those occurring in nitric acid and adipic acid plants. A commercial plant of this process is already available. (Uhde GmbH, Dortmund, Germany)

The state of iron in Fe-ZSM-5 is strongly dependent on the method of iron exchange, the level of Fe exchange (i.e., the Fe/Al ratio), and the pretreatment of the as exchanged material. In general, it is agreed on that tetragonally coordinated iron ions in the zeolite framework as well as iron oxide particles at the external surface of the zeolite crystal are inactive for N2O decomposition. On the other hand, it is still controversial if isolated single iron sites or diferric oxo/hydroxyl-bridged clusters are the active site for the N2O decomposition.

The reaction mechanism of the nitrous oxide decomposition has been studied on hydrated and dehydrated mononuclear iron sites in Fe-ZSM-5 using density functional theory.[1] In total 46 different surface species with different spin states (spin multiplicity MS = 4 or 6) and 63 elementary reactions were considered. Heats of adsorption, activation barriers, reaction rates and approximate minimum energy pathways were determined. Spin surface crossing (e.g. O2 desorption) was considered.

Based on these elementary ab initio rate constants several steady state and transient-response experiments for the decomposition of N2O over Fe-ZSM-5 have been simulated from first principles and compared with experimental results.[2] It is shown that a reaction mechanism based on single iron sites (viz., Z-[FeO]+) provides a satisfactory basis for simulating all of the experimental work reported in the literature. The overall rate of N2O decomposition is found to be first order in N2O partial pressure and zero order in O2 partial pressure. The presence of H2O in the feed gas inhibits the rate of N2O decomposition by deactivating the active sites required for this process. Site deactivation occurs via the process Z-[FeO]+ + H2O Z-[Fe(OH)2]+, which is reversible, and consequently, the influence of water vapor is strongly temperature dependent. Because of the high activation barrier for H2O desorption, dehydration of a hydrated sample of Fe-ZSM-5 can take as much as 10 h at 773 K. The presence of low concentrations of water vapor in the feed stream (ppb to ppm levels) affects the measured values for the apparent activation energy and preexponential factor, and this explains the large spread in values reported in the literature, as well as the appearance of an apparent compensation effect. Finally, it is shown that the desorption of O2 observed at ~ 900 K in TPD experiments conducted after the use of Fe-ZSM-5 for N2O decomposition is due to the process Z-[FeO2]+ Z-[Fe]+ + O2; however, this process is not kinetically relevant during steady-state decomposition.

[1] Heyden, A., Peters, B., Bell, A. T., Keil, F. J., J. Phys. Chem. B. 109, 1857 (2005).

[2] Heyden, A., Bell, A. T., Keil, F. J., J. Catal., in press.