287523 Influence of Pt Dispersion On NO Oxidation for the Modeling of Lean NOx Trap Catalyst Degredation

Wednesday, October 31, 2012
Hall B (Convention Center )
Travis Wentworth, Chemical & Petroleum Engineering, University of Kansas, Lawrence, KS, Susan Williams, Dep't of Chemical & Petroleum Engineering, University of Kansas, Lawrence, KS, Christopher D. Depcik, Mechanical Engineering, University of Kansas, Lawrence, KS and Loya Sudarshan, Mechanical Engineering, University of Kansas, Lawrence

Recently the implementation of lean burn diesel engines for commercial and consumer use has been increasing in popularity due to their increased fuel efficiency.  These engines generate higher NOx concentrations than traditional stoicheometric engines.  Stringent EPA regulations on the emissions of lean operating diesel engines has necessitated a major research emphasis in the area of NOx reduction due to the ineffectiveness of traditional three way catalysts to convert these exhaust gases. 

This paper focuses on a lean NOx trap (LNT) catalyst produced by incipient wetness impregnation of gamma alumina (Strem), to yield the desired weight percentage loading.  The catalysts examined and corresponding wt. % loadings are as follows: Pt/Al, Pt/Ba/Al, Pt/Ba/Ce/Al catalysts with weight percent loadings of 1/100, 1/20/100 and 1/20/5/100 respectively.  Pt particle size and dispersion have been estimated through CO chemisorption and verified by TEM.  NO and total NOx concentrations have been measured through Chemiluminescence detection.  

A fundamental understanding of the deviation of the oxidation step with catalyst degradation and the ability to efficiently model this phenomenon is a very important step in the implementation of LNT catalysts as engine exhaust after treatment devices.  The NO oxidation light off curves is presented for each of the catalyst formulations with varying platinum particle dispersion, in the range of 50-500°C. The experimental values obtained will be used to estimate model parameters for a computationally efficient 1-d model.


Extended Abstract: File Not Uploaded