| Resumo: | In the crystal form of rutile or anatase, the TiO2 surface is one of the most studied metallic oxide surfaces, due to its potential applications in the catalysis field, in addition to other technological uses. The reactivity of the TiO2 surface is governed in part by its concentration of defects, such as oxygen vacancies, with its catalytic activity also being affected by the adsorption of atmospheric gases such as water or oxygen. In the preceding work performed by the Applied Surface Science group of CEFITEC, the kinetics of water adsorption on the TiO2(110)-(1×1) surface were studied by monitoring the time-dependence of the work function change during the adsorption process, by measuring the cut-off position of the secondary electron energy spectra (onset method). This technique has proven to be a very useful tool for the analysis of adsorption dynamics, due to its high sensitivity to adsorbates which modify the surface dipole layer, and short acquisition time requirement. In the work here presented, this study is extended, and the onset method is used as a primary tool, together with XPS, for performing an analysis of the adsorption dynamics of water and oxygen on the rutile TiO2(110)-(1×1) surface, and how this process is affected by the surface temperature, and the presence of surface defects or contaminants. First, the physical model behind the onset method was revised, as some of the previous results obtained with this technique could not be explained under the present interpretation of the measurement. In this revision, the electric field between surface regions with different local work functions is considered, and its effect on the onset position is calculated as a function of the surface morphology and external fields. The TiO2 related experiments presented in this thesis were performed on a Kratos XSAM 800 system. Before initiating the experimental campaign, the Kratos system was upgraded to the needs of this work. Its control and acquisition hardware system and software was replaced, so that both control and data acquisition tasks could be performed with a modern PC. The spectrometer was recalibrated by measuring its transmission function with two methods, the standard first principles method, and a newly introduced biasing method, which presents clear advantages with respect to the first. Lastly, a method was introduced for the optimization of the transmission function, by a applying the differential evolution search algorithm to the voltages of the spectrometer’s electrodes, resulting in an increase of the system’s sensitivity in the whole energy range. Finally, the onset method was used to study the presence of surface defects and the adsorption of water and oxygen on the rutile TiO2(110)-(1×1) surface. A water adsorption model was introduced which was capable of explaining the results that were previously obtained. The adsorption of hydrocarbon contaminants, accumulated in the UHV system over time, was found to be of substantial importance. It was possible with this model to calculate the constants of the adsorption kinetics and estimate the concentration of water molecules adsorbed on Ti rows and dissociated on bridging oxygen vacancies. The oxygen adsorption process was monitored both by work function and XPS measurements. The adsorption kinetics were determined for surfaces prepared with different reduction levels. Oxygen adsorption was also associated with a decrease of the Ti3+ contribution in the XPS Ti 3p peak, and the magnitude of this decrease was linearly correlated with the increase of the work function following oxygen adsorption. The process of oxygen adsorption were found to be influenced by the presence of both bridging oxygen vacancies and Ti interstitial species. The onset method was also used as a tool to study the electron stimulated desorption of oxygen and hydrogen species from the TiO2(110)-(1×1) surface. A Monte-Carlo simulation was used to estimate how the cross-section of oxygen species is affected by the local the presence of oxygen vacancies, and its value was determined for bombardment energies of 40 and 80 eV. The ESD cross-section of hydrogen species for a bombardment energy of 30 eV was measured by first exposing the surface to water, thus introducing surface OH groups, and then initiating the electron bombardment.
|
|---|