Carbon dioxide adsorption and hydrogenation on nickel-based surfaces: a first principles study
The chemistry of carbon dioxide has recently become of great interest both for technological and environmental issues. Indeed, carbon dioxide is one of the most problematic greenhouse gases and is also a fundamental ingredient for the industrial catalytic organic synthesis of many compounds like, for example, methanol. Recent investigations have shown that, while the common industrial process for methanol synthesis is carried out on Cu catalysts, Ni-Cu model catalysts show a particularly high e ciency. In order to understand the origin of such increase in the catalyst activity, a thorough characterization of the CO2-Ni interaction and the atomic-scale description of the hydrogenation process are mandatory. This thesis is focused on the study of the adsorption and activation of carbon dioxide mainly on pure Ni(110) surface and of its reactions with atomic and molecular hydrogen by means of accurate quantum mechanical rst principles numerical simulations. The interaction of the CO2 molecule with the surface is characterized in terms of adsorption geometries, energetics, vibrational and electronic properties, including charge transfer, core-level shifts and scanning tunneling microscopy images, obtained from electronic structure calculations and compared with original experimental results achieved mainly by the Surface Structure and Reactivity Group active at the TASC laboratory. A consistent picture of CO2 chemisorption on Ni(110) is provided on the basis of the newly available information, yielding a deeper insight into the previously existing spectroscopic and theoretical data. We nd that CO2 molecule can be chemisorbed in diff erent, almost energetically equivalent adsorption confi gurations on a Ni(110) surface,with high charge transfer from the substrate. The molecule, that in gas phase is linear and unreactive, is chemisorbed in a bent and activated state on the nickel surface and can react with the hydrogen. The atomic-scale investigation sheds light also on the long-standing debate on the actual reaction path followed by the reactants. Di fferent hydrogenation channels have been explored to determine the reaction network: using molecular hydrogen, only a Langmuir-Hinshelwood process (both reactants are adsorbed) is possible, resulting in the production of formate which is just a 'dead-end' molecule; with atomic hydrogen, instead, the reaction proceeds also through parallel Eley-Rideal channels (only one of the molecules adsorbs and the other one reacts with it directly from the gas phase, without adsorbing), where hydrogen-assisted C-O bond cleavage in CO2 yields CO already at low temperature.