Elucidating the Mechanism of NOx Storage and Reduction

Michael  Harold, University of Houston

NOx storage and reduction (NSR) is a complex catalytic process involving periodic operation, multi-functional catalysts, multiple products, and strong coupling between adsorption, reaction, and transport processes.  Bench reactor experiments reveal the existence of traveling concentration and thermal fronts.  In order to carry out a rational design of the lean NOx trap (LNT), a comprehensive understanding of the reaction pathways and kinetics is essential.  Towards this goal, we are conducting a combination of atmospheric flow reactor experiments and high vacuum pulsing experiments in a TAP (temporal analysis of products) reactor to build a predictive microkinetic model for NSR. 

The TAP experiments employing isotopically labeled reactants provide insight about reaction pathways on the multi-functional catalyst.  A typical experiment involved the pre-nitration of a Pt/BaO with 15NO so that the storage phase initially comprised Ba(15NO3)2.  Subsequent exposure of the catalyst to a sequence of H2/NO pulses enabled the identification of the source of the N2 product.  This experiment simulates the storage and reduction cycle in the LNT.  The results reveal more than two routes to molecular nitrogen.  The sustained production of 15N2 during the 15NO pulse suggests the direct decomposition of 15NO on clean Pt sites.  These sites are continuously cleansed by scavenging of adsorbed O during the H2 pulse, and do not involve N species pre-stored on the BaO storage phase.  On the other hand, both 15NN and N2 are produced during both the 15NO and H2 pulses.  Their production clearly shows the involvement of NOx species that spillover from the Ba phase to the Pt.  A third source of N2 is from NH3 formed during the H2 pulse (H2 being in excess).  Additional data provide insight into the extent of gradients in the vicinity of the Pt crystallites.

Bench-scale reactor experiments were conducted to elucidate the role of Pt dispersion on the N-product distribution.  The high dispersion catalyst (50%) leads to higher rates of NO oxidation and stored NOx reduction with high selectivity to N2.  On the other hand, the low dispersion catalyst (3%) leads to significantly higher ammonia selectivity at high NOx conversion.  These findings, which have implications for the design of hybrid LNT-SCR units, will be discussed in terms of a phenomenological model of the spatio-temporal behavior of the LNT. 

A predictive LNT model has been developed that contains a microkinetic description of the NO/H2/O2 reaction system and NO oxidation and NOx storage processes, and that is consistent with the TAP findings.  Comparisons will be made to selected model predictions and experimental results.

Download Presentation: