Abstract:
The first part of this work focuses on gas reception on pristine SnO2, namely the nature of the active oxygen and electron donor species, as well as the interfering effect of water vapour. By using operando spectroscopies, namely DRIFTS and UV/vis-DRS, this work demonstrated that in dry and humid air or in the absence of oxygen the same surface species are involved in the reception of target gases, namely lattice oxygen. The electronic properties of pristine SnO2 are controlled by the surface concentration of oxygen vacancies, which act as an electron donor. The situation at the pristine SnO2 surface is determined by the interplay of reduction and (re-)oxidation reactions, which depends directly on the atmospheric composition. The effect of water vapour on pristine SnO2 was further studied by in-situ and operando DRIFTS at different operation temperatures. On the studied pristine SnO2, three regions with different dominant water adsorption processes were found depending on the operation temperature, namely physisorption at temperatures below 100 °C, associative adsorption of water up to 300 °C and dissociative adsorption above 300 °C; the transition between associative and dissociative depends on the properties of the SnO2 surface. The strongest effect of water vapour on the sensor resistance correlates with associative adsorption and decreases with increasing associative adsorption. In conclusion it is assumed, that associative adsorbed water decreases the resistance, either direct as an electron donor or indirect by hindering oxygen adsorption, while dissociative adsorption is charge neutral.
The second part of this work is dedicated to the impact of Pt on the gas reception and transduction mechanism of SnO2. As a first step, the structure of the Pt loadings was investigated using XAS and TEM. The Pt loadings form small Pt oxide clusters, which are well dispersed and in close contact with the SnO2 surface. As shown by CO oxidation measurements, the presence of Pt strongly enhances the reactivity of the material and, unlike undoped SnO2, sustains a high level of CO oxidation in humid air. However, only in humid air this increased reactivity results in higher sensor signals than on pristine SnO2. The surface chemistry the Pt-doped SnO2 samples were studied by operando DRIFTS. In dry air, the Pt oxide clusters are easily re-oxidized and the oxidation of CO does not change the surface, i.e. there are no changes that can be transduced into a sensor signal. In humid air, it was found that the oxidation of CO causes clear changes of the surface of both oxides - Pt oxide clusters and SnO2 - causing a change of the electronic properties of the surface, i.e. a sensor signal. Furthermore, it is found, that the Pt oxide clusters and SnO2 are electronically coupled. This electronic coupling, determines the space charge layer, i.e. it is no longer determined by the concentration of oxygen vacancies on SnO2 but rather by the electronic coupling of the two oxides, which depends on the composition of the Pt oxide clusters. In conclusion, the found effect of the Pt dopant is the one described by the Fermi-level control model.
In sum the presented results on undoped and Pt-doped SnO2 show a clear correlation of the reactivity of the material and the obtained sensor signals. If the material is not reactive, e.g. as observed on pristine SnO2 in humid air, the lack of an interaction prevents any gas reception and thus any sensing effect. If the material is very reactive, e.g. in the case of Pt-doped SnO2 in dry air, the material is a typical catalyst, i.e. it facilitates the reaction but is not changed after completing the reaction. Although there is a very strong interaction of the target gas with the surface, the surface remains almost the same and consequently there are no changes that can be transduced into a sensor signal. In conclusion, a suitable sensor material should have a moderate reactivity. The surface reaction has to cause temporary, i.e. fully reversible, changes of the surface, which are transduced into a sensor signal. Such situations are found in the case of undoped SnO2 in dry air or Pt-doped SnO2 samples in humid air, i.e. the suitability of a material is not only determined by its intrinsic properties; in fact, the specific conditions in the environment and the resulting reactivity of a material determine its sensing performance.
