(501d) Understanding Nox Storage on Pt/Bao/Al2o3 Catalysts Using Simulated Diesel Exhaust

Authors: 
Cumaranatunge, L., Purdue University
Mulla, S., Purdue University
Ratts, J. L., Purdue University
Cao, L., Purdue University
Kromer, B. R., Purdue University
Delgass, W. N., Purdue University
Ribeiro, F. H., Purdue University
Yezerets, A., Cummins Inc.


The focus of this research is to develop a model to describe the NOx storage ?reduction (NSR) process for decreasing the hazardous emission of oxides of nitrogen (primarily NO and NO2, collectively referred to as NOx) from vehicle exhaust for lean-burn engines. The NSR process can be used in engines that operate in excess oxygen for fuel combustion, known as lean-burn engines (e.g., diesel engines) to efficiently reduce NOx in engine exhaust to harmless N2. The presentation details the experimental and modeling efforts carried out to understand the NOx storage step on a model NSR catalyst using simulated diesel exhaust. A 0.4wt% Pt/ 20wt% BaO/ Al2O3 monolith has been utilized in a plug flow reactor configuration at a space velocity of 32000 h-1. The exit gases were analyzed using an FT-IR analyzer. Capture-regeneration cycles up to the point of catalyst saturation (i.e., complete NOx slip) have been carried out to determine the effect of various gas compositions (NO, NO2, O2, H2O, CO2) at varying temperatures (150-350 oC). A significant difference in the shape of the NOx uptake curve is seen at 150 oC compared to higher temperatures (200-350 oC) while the adverse effects of CO2 and H2O on NOx storage seem to depend on the source of NOx (i.e., NO or NO2). Varying the CO2 and H2O concentrations in the 5-13% range does not seem to have any effect on storage. The presence of CO2, under dry conditions, in a NO2 + O2 feed seems to decrease the total NOx storage by 11- 15% while the presence of H2O (with NO2 + O2 and without CO2) seems to decrease the storage by only 2-3 %. This has been attributed to NOx being stored on barium hydroxides more easily than on barium carbonates. Finite-element reactor modeling have shown that NOx uptake on these catalysts is mass transfer limited, and at least two different sites, fast and slow uptake sites, are necessary to describe the experimentally observed NOx capture curves. Two probable models have been proposed including; (1) a series-site model, involving fast storage on the surface sites and slower storage through diffusion of surface species into the bulk of the sorber material, Ba and (2) a parallel-site model, involving fast uptake on Ba in proximity to Pt and slower uptake on isolated Ba particles. Capture experiments with NO in the absence of O2, CO2 and H2O in the feed, have shown that NO stores significantly (up to 50%) compared to NO2 while the presence of CO2 and H2O decreases NO storage to about 12% that of NO2. The addition of O2 to the feed with NO increased storage significantly. An additional storage site, for the storage of NO, has been added to the model in order to match the NO evolution curve seen experimentally during NOx storage for a NO2 + O2 feed. The presence of H2O seems to significantly decrease the storage of NO while the storage of NO2 remains unchanged. This emphasizes the need to have a good NO oxidation component in these catalysts. The effect of barium and platinum loadings on storage will be discussed.