(755b) Nature of Active Sites for Passive NOx Adsorption on Pd-SSZ-13
AIChE Annual Meeting
Friday, November 15, 2019 - 8:21am to 8:42am
Pd-exchanged SSZ-13 with CHA framework has shown great promise in trapping NOx at temperatures below 150 °C and releasing them above 200 °C. With the help of density functional theory (DFT) simulations combined with experimental techniques including temperature programmed desorption (TPD) and temporal analysis of products (TAP) on 1% Pd-SSZ-13, we have investigated the nature of these active sites. Based on their transient responses to temperature and feed changes, we propose a viable mechanism of elementary reaction steps describing to the trapping and release of NOx.
Our DFT results indicate that NO adsorption is the strongest on the partially reduced ZPdI sites, with a strength of ca. -250 kJ/mol. While the presence of H2O weakens the adsorption strength to -200 to -220 kJ/mol, NO adsorption remains relatively strong. The ZPdI sites, however, are not as stable as the ZPdIIOH or Z2PdII sites, suggesting that ZPdI is a minority site that may dynamically form in response to temperature or feed changes. We investigated the formation of ZPdI sites and considered a two-step reduction mechanism converting two adjacent ZPdIIOH sites to ZPdI sites. In the first step the adjacent ZPdIIOH sites dehydrate to form the bridged Z2[PdII-O-PdII] species, which is subsequently reduced in the presence of NO to form two ZPdI sites. With NO as the only reductant the formation of a ZPdI site from ZPdIIOH is endothermic by ca. 70 kJ/mol, but in the presence of stronger co-reductants, such as ethylene or CO, the energetics become more favorable: ethylene lowers the energy required to reduce ZPdIIOH to ca. 10 kJ/mol and the even stronger reductant CO renders the formation of ZPdI exothermic by ca. -35 kJ/mol.
Our TPD results for 1wt% Pd-SSZ-13 (Si/Al ~12-25) with NOX uptake at 100 °C followed by desorption with a temperature ramp of 20 °C /min indicate good NO uptake with a NO:Pd ratio of ~1 in presence of water. Higher uptake is possible by adsorption on Brønsted acid sites, which become available when water is removed from the feed. The presence of other reductants, such as CO or ethylene, also leads to a significant increase in NO uptake. These observations are in good qualitative agreement with the conclusions drawn from our DFT calculations. TAP analysis using isotope labelling studies in the temperature range of 100-450 °C with 16O2 pretreated catalyst,pulsed with 15N18O, indicates the formation of 15N16O18O.The incorporation of the labelled oxygen in NO2 is consistent with the second step in the two-step mechanism proposed in DFT calculations.
Our detailed mechanistic insight into the elementary steps describing the dynamic changes of active sites during NO trapping and NO2 production suggests that the relative stability of reduced, oxidized and hydrated/hydroxylated cationic sites is paramount. This knowledge provides the basis for future improvements of PNA materials and ultimately, cleaner automotive emissions.