Title : Hydrogen evolution over Gallium oxynitride prepared from Gallium oxide hydroxide under visible light irradiation
Abstract:
We study water splitting reactions using solid photocatalysts. It is well known that only 6% or less energy of sunlight is given by UV light, while more than 50% by visible light. Therefore, it is important to prepare photocatalysts driven by the visible light for the efficient use of the sunlight energy. Recently, it has been reported that nitrogen doping into metal oxides contributes to narrowing of their band gap, providing a visible-light response because N 2p orbitals are newly formed above O 2p orbitals in their valence band. To generate visible light response in gallium oxide (Ga2O3) photocatalysts, which showed high photocatalytic activities of water splitting reaction under UV light irradiation, we have tried nitrogen doping into Ga2O3. However, under visible light irradiation, gallium oxynitride (referred as GaON) prepared from Ga2O3 did not show the activity even for hydrogen evolution from methanol aqueous solution. In this study, GaON samples were prepared by calcination of gallium oxide hydroxide (GaOOH), a precursor of Ga2O3, under NH3 atmosphere at 773-1173 K. The prepared photocatalysts were characterized by Field Emission-Scanning electron microscopy (FE-SEM), UV-vis diffuse reflectance (DR), powder X-ray diffraction (XRD), X-ray absorption fine structure (XAFS) and X-ray photoelectron (XP) spectroscopies to analyze their optical properties and crystalline structures.
N K-edge XANES spectra showed that prepared GaON samples contained nitrogen after nitrogen doping. The spectral profiles represented the chemical states of doped nitrogen, i.e., molecular nitrogen existed as interstitials when nitrogen doping temperature was under 973 K and above 973K gallium nitride (GaN) precipitated. However, radial structure functions of Ga K-edge EXAFS spectra of the samples were not strictly consistent with that of commercially available GaN, suggesting that some Ga-O bonds is remained. H2 evolution reactions from methanol aqueous solution using prepared samples were performed under visible light irradiation (λ>420 nm). This reaction did not proceed over commercially available GaN and prepared samples except the sample prepared at 1173 K. In order to understand why such difference in photocatalytic activity appeared, we investigated XRD patterns before and after the reaction. We found that bulk crystalline structure of the sample prepared at 1173 K fundamentally maintained while those of others had changed probably due to release of doped nitrogen. In this conference, we would like to discuss this topic in detail and ask you to give some advises to us.
