The optimization of the synthesis of Ni/Sb-SnO2 nanopowders for the electrochemical generation of ozone

Abstract

Nickel and antimony co-doped tin oxide (Ni/Sb – SnO2) anodes have been shown to exhibit high selectivity for ozone generation at current efficiencies of ca. 50% and current densities of ca. 100 mA cm-2 . However, the synthesis involves dip coating Ti substrates in solutions containing the precursors, calcining at temperatures between 400 and 900 o C, and repeating these steps a number of times to produce ceramic coatings with geometric surface. The synthesis proved to be somewhat irreproducible, and the current densities that could be attained were low. The latter precluded the use of these anodes to treat “real” waters in the absence of added electrolyte: such a process would require a zero gap, polymer electrolyte membrane (Nafion) cell. Unfortunately, at 100 mA cm-2 there is a problem of poisoning polymer electrolyte membrane by cations such as Na+, Mg2+, and Ca2+ present in such waters, thereby rendering the ceramic anodes incapable of treating real waters. The postulate that forms the basis of the work reported in this report was that if high surface area, active and selective nanopowders of Ni/Sb – SnO2 could be synthesised then the current densities could be sufficient to prevent fouling of the polymer electrolyte membrane and allow the application of zero gap cells in water treatment. Preliminary work in Newcastle had suggested that hydrothermal method could be employed to produce active materials, in which the precursors salts were dissolved in deionized water and refluxed before the hydrothermal process and calcination. The second method explored in this thesis was the same as the first except that Sb-SnO2 (ATO) was synthesized, calcined, and then doped with Ni by mixing the powder with the appropriate weight percent of nickel chloride solution before calcination. It was hoped that this latter method would be more reproducible, as well as producing active and selective Ni/Sb – SnO2 nanopowders. The nanopowder samples prepared from the two routes were characterized using X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), Energy dispersive X-ray (EDX), and Brunauer Emmett-Teller to evaluate how much the changes in compositions and synthesis method affected some properties of the nanopowder samples. The XRD results revealed that increasing the Sb content of the nanopowder samples inhibited crystal growth and reduced the crystallite size, whereas increasing the amount of Ni at fixed iii Sb content and calcination temperature had no effect on the crystallite size. It also showed that the method of doping Ni had little or no effect on crystallite size. The BET analysis revealed that surface areas of the nanopowders from both routes increased as the Sb concentration increased. Cyclic voltammetry was employed to ascertain the onset potentials and potential ranges for redox processes. In general, the cyclic voltammograms of the ATO and the Ni/Sb – SnO2 electrodes in acidic electrolytes showed redox peaks. The responses of the ATO and the Ni/Sb – SnO2 anodes were similar; hence the peaks were attributed to the redox process associated with the Sb and Sn at the surface. The presence of the peaks in the cyclic voltammograms of these electrodes is most likely because the electrodes were prepared with high surface area powders which rendered the features visible in the voltammetry, unlike that of the ceramic electrodes geometric areas reported in the literature. Ozone activities and selectivity of the Ni/Sb – SnO2 nanopowders prepared by both routes were investigated in 0.5 M H2SO4 using a UV-Vis cuvette as the electrochemical cell. Ni/Sb – SnO2 anode was held flat on the bottom of the cuvette containing the electrolyte, and a 0.64 cm2 Pt/Ti mesh cathode inserted was vertically. The electrodes were placed against the opaque sides of the cuvette to avoid interfering with the light passing through the UV Vis spectrometer. On comparing the data obtained, it was found that the first doping route where the precursors salts were dissolved together before the hydrothermal process is more active in ozone generation. In general, it was observed that ozone activity decreased at higher Ni concentrations for nanopowders of both routes. XPS analysis of the nanopowders showed the presence of Sn at the oxidation state of +4 and Sb at the oxidation state of +3 and +5. Ni was not detected, probably because of the detection limit of the XPS or the location of the Ni in the lattice. On comparing the Sb/Sn atom ratio of the Ni/Sb – SnO2 nanopowders from the two routes, it was found that the second doping route had higher surface enrichment of Sb3+ than the Ni/Sb doping route, and this could be the reason for the observed high ozone current efficiency with the NATO anodes. The effects of ozone on the skin using the Phenion Skin Model were investigated. Three skin models were exposed to 0.3 ppm ozone in the exposure chamber for a week at 8hr each day, and three were kept as control. Haematoxylin and eosin (H&E) staining was employed to iv determine any changes to the thickness of the skin layers due to ozone exposure. In brief, it was found that ozone exposure increased the thickness of the epidermis and decreased the thickness of the dermis. The effects of ozone on the protein content of the skin models were also analysed using the Bradford assay, and it was observed that ozone caused the release of protein from the skin cells models.