Artificial neural network-aided design of Co/SrCO3 catalyst for preferential oxidation of CO in excess hydrogen

Preferential oxidation (PROX) of 0.7–1 vol.% CO using the stoichiometric amount of O2 was investigated in excess hydrogen. Cobalt loading and preparation conditions of Co/SrCO3 was optimized by using a full factorial design of experiment, an artificial neural network and a grid search. The optimum catalyst was 3.2 mol% Co/SrCO3 pretreated at 345 °C and 97% CO conversion was achieved at 240 °C under dry and CO2 free conditions. However CO2 and H2O vapor inhibited the activity, and the new additive to the Co/SrCO3 catalyst was investigated in the next step for the high tolerance towards CO2 and H2O. Representative 10 elements (B, K, Sc, Mn, Zn, Nb, Ag, Nd, Re, and Tl) were selected to represent the physicochemical properties of all elements. Based on the relation between the physicochemical properties of element X and the catalytic performance of Co–X/SrCO3, the elements such as Bi, Ga, and In were predicted to be promising additives. Finally, the catalytic performance of these additives was experimentally verified. Sixty-four percent CO conversion and 70% selectivity for PROX at 240 °C was achieved in the presence of excess carbon dioxide and steam by Co 3.2–Bi 0.3 mol%/SrCO3 pretreated at 345 °C.