Hydrogen production from oxidative reforming of methane on supported nickel catalysts: An experimental and modeling study

► Autothermal reforming of methane was studied using nickel based catalysts. ► A kinetic model from the literature was used to compare the obtained results. ► Catalysts containing alumina presented better metallic dispersions. ► Higher dispersions led to superior methane conversions and H2 yield the reaction. ► Literature model predicted well the general behavior of the reaction. Nickel catalysts supported on Al2O3, CeO2/Al2O3, Ce0.5Zr0.5O2/Al2O3, and Ce0.5Zr0.5O2 were investigated for methane oxidative reforming. BET surface area results showed that the catalysts containing alumina presented higher surface area which favored better nickel dispersion. This was confirmed by X-ray diffraction (XRD) data from the reduced samples. XRD analysis of the calcined catalysts also showed the formation of a ceria–zirconia solid solution. In fact, the addition of CeO2 and CeZrO2 to alumina provided higher oxygen storage capacities as observed by CO2 formation during CO-TPD. Temperature Programmed Reduction and Diffuse Reflectance Spectroscopy experiments revealed that samples containing alumina showed higher interaction between metal and support. Samples supported on alumina showed similar methane conversions during oxidative reforming, which could be related to similar nickel dispersions. A literature based kinetic model was used to compare data predicted by this model with the experimental behavior. The model results of methane conversions and composition profiles indicated that high temperatures should be used in order to obtain a maximum in H2 production. The model predicted smaller methane and oxygen conversions, as well as, lower H2 and CO molar fractions than the ones observed experimentally. This happened probably due to the fact that the kinetic expressions used were obtained at lower temperatures, lower conversions and with a different catalyst. Despite these differences, the general behavior was predicted by the model.