(7b) Au/CeO2 Catalysts for Water Gas Shift Reaction

Thanks to their high energy efficiency and low emissions, PEM-FCs are the candidates for electric power generation in mobile applications. The most probable fuel is H2-rich gas coming from reformed hydrocarbons; at reformer outlet it contains about 10% CO. The performance of Pt, used as catalytic element for H2 oxidation in PEM-FCs, is seriously depressed from CO; therefore the latter must be removed through water gas shift (WGS) reaction (to about 0.5?1% with simultaneous increase of H2) and then below 10 ppmv by CO preferential oxidation (CO-PROX) or selective methanation (CO-SMET). Gold supported catalysts have revealed extremely active at low temperatures in different reactions when highly dispersed on suitable carriers: dispersion strongly influences activity and selectivity. An efficient deposition method able to guarantee an Au intimate dispersion on the carrier is thus a decisive issue to provide good catalytic performance. The goal of the present work is to evaluate the influence of synthesis parameters on Au dispersion on ceria support and the activity toward CO clean-up towards the medium temperature (MT) WGS reaction. CeO2 was prepared through the solution combustion synthesis (SCS) technique followed by calcination in calm air at 650 °C for 2 h to obtain crystalline structure. Au was added by deposition-precipitation method, trying to obtain the smallest particle size, necessary for a satisfactory catalytic activity. An aqueous solution of the Au precursor (HAuCl4) was prepared adjusting the pH to the desired value and increasing the temperature to 80 °C; CeO2 powder was then added, stirring on a heated plate 1 h. The suspension was then filtered, washed and dried under vacuum. Different catalysts were prepared, varying both pH (5, 7, 8.5 and 10) and the molarity M of precursor solution (0.2?10-3, 0.5?10-3, 0.73?10-3 and 1?10-3) for each of them the same calcination conditions were employed: calm air at 400 °C for 3 h. The as-prepared Au-based catalysts were then characterized by X-ray diffraction analysis and field emission electron microscopy. The BET surface area and porosimetry distribution of the catalysts was determined by means of N2 adsorption with an automated gas adsorption analyzer and the metal dispersion and distribution by chemisorption. A fixed bed micro-reactor (a quartz tube of 4 mm I.D.), heated up by a PID regulated oven and containing 0.3 g of catalyst in powder, diluted with 0.5 g of SiO2 (0.2?0.7 mm), held in place by flocks of quartz wool, was used for the catalytic activity tests. A K-type thermocouple was inserted into the catalytic bed to measure its temperature. The catalysts were firstly reduced into the micro-reactor flowing 50% H2 in He (100 Nml min-1) at 200 °C for 1 h, and then tested between 100 and 450 °C, with a weight space velocity (WSV) of 0.333 NL min-1 gcat-1. The best catalysts were tested also at a WSV of 0.167 and 0.5 NL min-1 gcat-1 As preliminary screening of the catalysts performance, various inlet gas compositions were fed to the reactor: first a mixture composed of 5% CO, 20% H2O and N2 as balance, then 5% CO, 20% H2O, 11% CO2 and N2, then 5% CO, 20% H2O, 40% H2 and N2, finally 5% CO, 20% H2O, 11% CO2, 40% H2 and N2. The outlet gas stream was analysed through a micro gas-chromatograph (Varian CP-4900) equipped with a thermal conductivity detector (TCD). The CO detection limit was 1 ppmv. The best catalytic activity in terms of CO abatement (with the he complete reactive mixture) was obtained with the sample prepared at pH = 8.5 and M = 0.2?10-3: with 89% CO abatement at 500 °C for the 2-components mixture and 12% CO abatement at 470 °C for the 4-components mixture at the intermediate WSV. The results are very encouraging. More details will be given in the full paper.