Chemical poisoning by zinc and phosphorous of Pt/Ba/Al2O3 NOx storage catalysts

Journal article, 2019

The effects of phosphorous and zinc on the performance of Pt/Ba/Al2O3 catalysts were investigated through wet impregnation of ammonium phosphate and zinc acetate aqueous solutions. Six different sample combinations were studied; 1 wt-% P, 1 wt-% Zn, 1 wt-% P with 1 wt-% Zn, 2 wt-% P, 2 wt-% Zn, 2 wt-% P with 2 wt-% Zn. NOx storage and reduction (NSR) activity and NO2 temperature programmed desorption (TPD) profiles were measured before and after impregnation of P and Zn. Samples containing P performed significantly worse than samples only containing Zn in both NSR activity and TPD measurements. The increased NOx slip during lean phase in activity measurements for P-poisoned samples is mainly related to an increased slip of NO2. This was found for both NO and NO2 in the gas feed during lean phase and suggests that it is mainly the storage component that is poisoned and not the noble metal. Furthermore, the combination (1 wt-% P and 1 wt-% Zn) proved to result in slightly worse performance than only 1 wt-% P, however this was not the case for samples containing 2 wt-%, where the addition of zinc reduced the negative effect of phosphorous. Measurements from NO2-TPD experiments showed that NOx release at low temperature was not affected by the addition of P, while desorption in the temperature range 425–475 °C was significantly reduced. It can therefore be concluded that the poisoning mainly is related to barium NOx storage sites and not to alumina sites. Moreover, X-ray diffraction measurements indicate that some of the barium species are affected by phosphorous. Images from scanning transmission electron microscopy (STEM) and energy-dispersive X-ray spectroscopy (EDX) mapping were in line with the results seen in both the activity tests and NO2-TPD experiments. Phosphorous was concentrated at the same position as barium in the observed images, whereas zinc was more evenly distributed over the surface. For the sample with both 2 wt-% P and 2 wt-% Zn, X-ray photoelectron spectroscopy measurements indicate that Zn and P have a low interaction and this suggests that most of the zinc and phosphorous are separated. However, STEM-EDX showed agglomerates of some zinc and phosphorous, which could be zinc phosphates. This is a plausible explanation of the decreased deactivation observed after introducing 2 wt-% Zn compared to the sample only containing 2 wt-% P.