Formation and removal of Ba-carbonates or carboxylates on Pt/BaO/ gamma -Al sub(2)O sub(3) lean NO sub(x) traps

• Published in: Applied catalysis. B, Environmental Vol. 107; no. 1-2; pp. 26 - 33
• Main Authors: Chaugule, Saurabh S, Kispersky, Vincent F, Ratts, Joshua L, Yezerets, Aleksey, Currier, Neal W, Ribeiro, Fabio H, Delgass, WNicholas
• Format: Journal Article
• Language: English
• Published: 31.08.2011
• Subjects: Adsorption; Carbon dioxide; Carbonates; Carboxylates; Nitrogen dioxide; Platinum; Regeneration; Surface chemistry
• ISSN: 0926-3373
• DOI: 10.1016/j.apcatb.2011.06.029

In addition to feed composition, the NO sub(x) storage capacity (NSC) of a Pt/BaO/ gamma -Al sub(2)O sub(3) lean NO sub(x) trap (LNT) also depends on the chemical state of Ba prior to the start of the lean NO sub(x) exposure phase during which the NSC is measured. This state of Ba is a result of the history of the trap's cyclic exposures to lean and rich conditions and is a function of the feed composition and the time intervals of the exposures. We have systematically investigated the role of adsorbed CO sub(2) on the NO sub(x) sorption process on traps with 20, 8 and 4 wt.% Ba. This was accomplished through a sequence of NO sub(x) storage and regeneration cycles with CO sub(2) and H sub(2)O in the feed followed by purge in Ar and then by NO sub(x) storage and regeneration without CO sub(2) and H sub(2)O at 300 degree C. The amount of pre-adsorbed CO sub(2) released during NO sub(x) storage upon the introduction of NO sub(2) + O sub(2) correlated well with the amount of NO sub(x) stored on the trap with 20 wt.% Ba but not on the traps with 8 and 4 wt.% Ba loadings. These results were further investigated in DRIFTS sequential adsorption experiments for CO sub(2), NO sub(2) + O sub(2), H sub(2) or NO sub(2) + O sub(2), CO sub(2), H sub(2) at 300 degree C. The DRIFTS experiments showed that, similar to bulk vs. surface nitrate formation on samples with high vs. low Ba loading, CO sub(2) adsorption forms either carbonates or carboxylates on high vs. low Ba loadings, respectively. Regeneration by H sub(2) removed all NO sub(x) adsorbates and any carboxylates, but not the carbonates. Thus, the formation of carbonates slows the NO sub(x) storage process as NO sub(x) must compete with pre-adsorbed carbonates for adsorption sites after regeneration. In addition, comparison between the NO sub(x) breakthrough curves with 7% CO sub(2) in the lean feed and with pre-adsorbed CO sub(2) on the catalyst showed that the NO sub(x) storage before the slip occurs was intimately related to the oxygen spillover from Pt to the surrounding Ba and competition between CO sub(2) and NO sub(x) for adsorption sites. These findings provide insights to further improve the LNT formulations by using optimum Ba loadings and dispersions targeted at increasing the zero NO sub(x) slip time after each fuel rich pulse.