Joshua L. Ratts1, Lasitha Cumaranatunge1, W. Nicholas Delgass1, Aleksey Yezerets2, Neal W. Currier2, and Fabio H. Ribeiro1. (1) Chemical Engineering, Purdue University, 480 Stadium Mall Drive, West Lafayette, IN 47907, (2) Catalyst Technology and Emissions Chemistry, Cummins Inc., 1900 McKinley Ave., MC 50183, Columbus, IN 47201
Operating internal combustion engines in a lean burn operation offers higher fuel efficiency when compared to a stoichiometric burn operation. Current three-way catalysts do not effectively reduce NOx under lean burn exhaust conditions so one method proposed to reduce NOx emission is through a cyclic operation of the engine between lean and rich periods and the use of an NOx Storage/Reduction (NSR) catalyst. During the lean period NOx is stored on the catalyst and it subsequently is reduced during the rich period. An in-situ DRIFTS study was conducted on a Pt/BaO/Al2O3 (1.31/9.0 wt%) catalyst which followed the time dependent surface species during typical lean/rich burn cyclic exhaust conditions. It was found that the catalyst needed to have many cycles run before reproducible results were obtained. This was due to the barium phase changing its state (e.g. carbonates, carboxylates, hydroxyls, nitrates). Once the surface was stabilized, simple mixtures of NO, NO/O2, NO2, and NO2/O2 were stored primarily as bulk barium nitrates, nitrates on a monolayer of barium, and aluminum nitrates with multiple bonding configurations at 300°C. After 15 minutes of storage, the catalyst required ~2 minutes to reduce the monolayer barium nitrates, aluminum nitrates, and 93% of the bulk barium nitrates while 10 minutes total were required to reduce 100% of the bulk barium nitrates with 1% H2. Our results show that the addition of 5% H2O primarily reduced the storage on the monolayer barium and alumina sites where as the addition of 5% CO2 primarily reduced the storage on the bulk barium sites. Quantification of the rest of these changes will be presented via peak deconvolution along with the time dependent evolution of each surface species. Flow reactor experiments were also carried out with the catalyst powder. The catalyst showed a typical breakthrough profile with 1.3 minutes of complete capture under the same conditions. Quantification of the amount of NOx reduced on the catalyst with different regeneration times agreed well with the in-situ DRIFTS data.