Krisztina Frey

Activity of SiO2 supported gold-palladium catalysts in CO oxidation

Abstract Bimetallic Au-Pd catalysts supported on silica with different Au/Pd atomic ratios were prepared by simultaneous reduction of palladium and gold precursors by ethanol in the presence of the polymer, polyvinylpyrrolidone (PVP). Formation of alloyed particles was detected by means of X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM) and CO chemisorption measurements. The catalysts were tested in the catalytic oxidation of CO using a plug–flow reactor. The CO conversion was determined as a function of temperature. The monometallic palladium and the palladium-rich catalysts behaved quite similarly and were the most active catalysts. A drastic reduction of the CO oxidation activity was observed for the 50/50 Au/Pd catalyst and for samples with increasing amount of gold.

Role of promoting oxide morphology dictating the activity of Au/SiO2 catalyst in CO oxidation

The interfacial interaction of gold nanoparticles deposited on either model SiO2/Si(100) or high surface area amorphous or mesoporous silica with minute amounts of promoter oxide like “active” FeOx, TiO2 and CeO2 has been discussed. The role of the active oxide, its contribution to the perimeter along the gold nanoparticles has been interpreted. The oxide may invoke electronic interaction and simultaneously the defect structure of oxides likely has a key issue in the formation and stabilization of very small Au particles. The activity of the Au/oxide perimeter depends not only on the size of the Au particles, but also on the size and morphology of the oxide component (likely amorphous structure) regardless of whether it is supporting Au nanoparticles or decorating them. The activity in CO oxidation over Au catalysts is strongly affected by the length of the Au/“active” oxide perimeter which is regarded as the “active interface”. The longer length of the perimeter is evidenced by the enhanced CO oxidation activity.

Reconstruction of low-count step-like signals in ion microbeam analysis

The evaluation of experimental data of ion microbeam analysis is often hindered by low counts/pixel statistics of elemental or tomographic images available within limited acquisition time, mainly at low elemental concentrations. In this contribution we present a method based on a non-linear iterative deconvolution scheme, designed to recover step-like profiles with forcibly low counts. To demonstrate its performance and effectiveness, the algorithm was applied for measured one-dimensional profiles with a micrometre resolution, generated by Rutherford backscattering spectrometry (μRBS) at the ATOMKI scanning nuclear microprobe facility. The method can be extended to solve the analogous two-dimensional problem, giving an improvement of nuclear microscopic methods, like channelling contrast microscopy (CCM).