Showing papers by "Louise Olsson published in 2011"

Mechanistic investigation of hydrothermal aging of Cu-Beta for ammonia SCR

Abstract: The selective catalytic reduction of NO with NH3 over a Cu-BEA catalyst was studied after hydrothermal aging between 500 and 900 degrees C. The corresponding catalyst was characterized using XPS and XRD techniques in the aging interval of 500, 700 and 800 degrees C. No structural changes during the aging process were observed. However, the oxidation state of copper changed during aging and more Cu2+ was formed. We suggest that one of the deactivation mechanisms is the decrease of the Cu+ species. The NO oxidation and NH3 oxidation activity was decreased with increasing aging temperature. Further, we observed that the ammonia oxidation was decreased faster compared to the SCR reactions at low aging temperatures. The experiments from the calorimeter as well as from the ammonia TPD investigations indicate a trend towards more weakly bound ammonia with higher aging temperatures. From the results of the SCR experiments using different NO2/NOx ratios and ammonia oxidation experiments we suggest that most of the N2O is coming from side reactions of the SCR mechanism and not from reactions between NH3 and O-2 alone. Interestingly, we observe that after the 900 degrees C aging a quite large activity remained for the case with 75% NO2/NOx ratio. The N2O production shows a maximum at 200 degrees C, but increases again at higher temperatures. However, the N2O formed at low temperature is decreased after hydrothermal aging while the high temperature N2O is increased. We propose that the different reactions examined in this work do not all occur on the same type of sites, since we observe different aging trends for some of the reactions.

Urea decomposition and HNCO hydrolysis studied over titanium dioxide, Fe-Beta and γ-Alumina

Abstract: The catalytic effect of titanium dioxide, Fe-Beta, gamma-Alumina, on the thermal decomposition of urea and hydrolysis of HNCO, was investigated using differential scanning calorimetry (DSC) and mass spectrometry (MS). The catalytic materials were coated on cordierite substrates and a pure cordierite sample was also used for comparison. The urea was administered by impregnating the monoliths with an urea/water solution. The experiments were performed using a constant heating rate of 10 K/min and over a temperature range of 25-500 degrees C. A sweep gas flow rate of 80 mL/min of either dry or humid Ar was used. The results show that TiO(2) is the best hydrolysis catalyst. Fe-Beta showed a very large ammonia production, due to selective adsorption of urea during the impregnation of the monolith in the urea solution. One experiment with lower flow, higher urea loading and increased ramp speed conducted in dry Ar over TiO(2) showed a large formation of biproducts. This experiment was repeated in the presence of water and this decreased the formation of CYA and biuret significantly. The reason for this is the effective hydrolysis of the HNCO over titania, which hindered the bi-product formation.

Kinetic modeling of NOx storage and reduction using spatially resolved MS measurements

Abstract: A Global Kinetic NOX Storage and Reduction (NSR) Model based on flow reactor experiments was developed to investigate the NOX storage and reduction mechanisms with a focus on the breakthrough of NH3 and N2O during the rich phase. Intra-Catalyst Storage and Reduction Measurements (SpaciMS) were used to further validate the model, particularly with respect to the formation and utilization of ammonia along the catalyst axis. Two different catalysts were used in the model, denoted Cat. 1 and Cat. 2. The first catalyst was used in flow reactor experiments to create a global kinetic model and fitting the parameters using long NSR cycles validated against more realistic short NSR cycles, while the second catalyst was used in the SpaciMS experiments. However, due to some differences in the catalytic material, some parameters had to be re-tuned for the second catalyst. Two NOX storage sites were used for both catalysts, barium (Ba) and the support sites (S2). Furthermore, the Shrinking-Core Model was used to describe the mass transport of NOX inside the storage particles, S2. An oxygen storage component was necessarily included and denoted Ce for the first catalyst and representing ceria in the catalyst. The second catalyst did not contain any ceria, which is why the oxygen storage site was called S3 and can be interpreted as oxygen on the noble metal. During the rich period, NOX was reduced by H2 and CO, forming nitrogen and NH3. Produced NH3 reacted with stored NOX forming N2O and resulting in an N2O peak before NH3 breakthrough. The model agreed well with reactor experiments and SpaciMS measurements. The SpaciMS results showed that most NOX was stored in the first half of the catalyst, resulting in high ammonia production in the catalyst front and its subsequent consumption along the catalyst axis to reduce NOX stored downstream.