Titaniumdioxide supported gold was found to catalyze the hydrolysis of formate-based ammonia precursor compounds which are proposed for the selective catalytic reduction of nitrogen oxides (NOx) in combustion exhaust gas. In contrast to other noble metals, the supported gold does not oxidize the released NH3, while it maintains decomposition of intermediate formic acid. Efficient formic acid decomposition is essential as it avoids the formation of side products which previously restricted the use of formate-based ammonia precursors.
The stability of the catalyst under operating conditions was investigated by an accelerated hydrothermal aging treatment and sulfur poisoning. Both procedures affected the catalytic decomposition performance only to a small extent, which is unprecedented for supported gold catalysts.
The novel catalyst was implemented into an up-scaled prototype system of an ammonia generator for automotive applications and successfully tested in an exhaust gas aftertreatment system of a 3.0 L Diesel engine on an engine test bench at the Technical University of Munich. The system was then transferred to an industrial company and is currently tested for commercial utilization.
The identification of the SCR active sites in Fe-ZSM-5 is of utmost importance for the understanding and optimization of the catalyst performance. No method (e.g. UV/VIS, IR, EPR, EXAFS, XPS, XRD) can definitively distinguish between isolated iron species and iron oxide clusters of different nuclearity in the same sample. A statistical approach was used to solve this problem. From the correlation of the measured SCR activity with the calculated concentration of different species the temperature dependent activities of isolated, dimeric and oligomeric iron species and iron particles could be determined.
Publications: S. Brandenberger, O. Kröcher, A. Tissler, R. Althoff, Estimation of the fractions of different nuclear iron species in uniformly metal-exchanged Fe-ZSM-5 samples based on a Poisson distribution, Appl. Catal. A 2009, accepted manuscript.
The hydrolysis of isocyanic acid was studied experimentally and theoretically and a reaction mechanism on different catalysts was established. The decreasing NOx emission limits for diesel vehicles impel the further development of the existing NOx deactivation technologies, particularly the selective catalytic reduction (SCR) of nitrogen oxides with urea. In the urea-SCR process, urea is injected into the hot exhaust gas, where it thermally decomposes into isocyanic acid (HNCO) and ammonia. HNCO quickly hydrolyses on the surface of SCR catalysts and even faster on the surface of specialized hydrolysis catalysts. The theoretical Density Functional Theory (DFT) method with cluster model was used. The reaction path was studied by adsorption of water and isocyanic acid, followed by optimization of the adsorbates structure. Additionally, vibrational analyses of the adsorbates were made. In the experimental part activity tests and surface sensitive techniques (XRD, XPS and DRIFTS) were applied. Measured DRIFT spectra could be reproduced by theoretical calculations. Based on the computational screening of different catalysts we can confirm the experimental results, which show that TiO2 is the best catalyst for isocyanic acid hydrolysis. Presented methodology of DFT modeling and "virtual" catalyst screening can be used for successful combination with different experimental methods available at PSI and for variety of questions and complex catalytic systems.