Development of Visible Light-Responsive Photocatalysts for Hydrogen Production from Glycerol
MetadataVis full innførsel
Different catalyst preparation methods were performed. In the case of TiO2, the cocatalyst was deposited on commercial TiO2-P25 by wet impregnation. Cu-loaded and Pt-loaded TiO2 was achieved by calcination in static air and H2 flow, respectively. Samples of SrTiO3:Rh was prepared by solid-state reaction (SS) and water-based hetero-chelate (WH) method. Cu and Pt-deposition on this photocatalyst was achieved by calcination in static air and in situ photodeposition in aqueous methanol solution, respectively. The prepared catalysts were then used in photocatalytic reactions using methanol and glycerol as organic substrates. Prior to activity measurements, the prepared catalysts were characterized with ultraviolet-visible spectroscopy (UV-Vis), X-Ray Diffraction (XRD), Nitrogen adsorption, Scanning electron microscopy (SEM), Transmission electron microscopy (TEM) and X-ray photoelectron spectroscopy (XPS). Activity measurements were done by irradiation of a 0.5 vol% aqueous solution of alcohol in a batch reactor. Initial runs revealed that glycerol was a better substrate than methanol, producing more hydrogen during the same irradiation period. Subsequent activity measurements were therefore solely focused on this substrate. XRD patterns confirmed anatase and rutile phases in the TiO2-P25 samples and perovskite SrTiO3 in the SrTiO3:Rh samples. Assignment of the crystal phase of the co-catalysts with XRD was not feasible due to low co-catalyst loading, but the deposition could be confirmed by TEM. Some impurity phases were identified for bare SrTiO3:Rh prepared via solid-state reaction, suggesting improper preparation. UV-Vis showed an increase in visible light absorption from co-catalyst deposition. Nitrogen adsorption showed that the heat treatment of bare TiO2-P25 samples would not change the BET surface area significantly. As for bare SrTiO3:Rh samples, the BET surface areas did not vary considerably. SEM images revealed mono-disperse particles for the SS samples. Although, WH samples exhibited smaller particles, they were not generally of uniform sizes. XPS confirmed the reduction to metallic Pt species after H2 reduction and photodeposition. It also showed that the pretreatment of Cu to obtain lower oxidation numbers was successful. Activity measurements showed that SrTiO3:Rh worked better in the visible region than TiO2-P25. Pt-loaded SrTiO3:Rh catalysts prepared by solid-state reaction yielded appreciable amounts of H2, whereas Cu-loaded SrTiO3:Rh samples had rather low activity even under UV irradiation. Unexpectedly, the WH sample hardly produce any evolved H2, strongly suggesting that the sample was not prepared as intended. Despite efforts to enhance photocatalytic activity of Cu-SrTiO3:Rh, the amount of evolved H2 evolution was not higher than the bare stoichiometric SS sample, bringing the need for an alternative deposition method or another cocatalyst. High Performance Liquid Chromatography (HPLC) detected the presence of glyceraldehyde in the reaction solution. Despite attempts to completely separate peaks attributed to glycerol and dihydroxyacetone, this was not feasible. Thus, selectivity and conversion calculations could not be done.