Mo-doped, Cr-Doped, and Mo–Cr codoped TiO2 thin-film photocatalysts by comparative sol-gel spin coating and ion implantation

Uniformly codoped anatase TiO2 thin films of varying (equal) Mo and Cr concentrations (≤1.00 mol% for each dopant) were fabricated using sol-gel spin coating and deposited on fused silica substrates. All films were annealed at 450 °C for 2 h to recrystallise anatase. Undoped anatase films have been subjected to dual ion implantation for the first time, using Mo, Cr, and sequential Mo + Cr at 1 × 1014 atoms/cm2. The films were characterised by GAXRD, AFM, SIMS, XPS, and UV–Vis and the performance was assessed by dye degradation. Despite the volumetric doping by sol-gel and the directional doping by ion implantation, neither method resulted in homogeneous dopant distributions. Both methods caused decreasing crystallinities and associated partial amorphisation. The XPS signal of the uniformly codoped films is dominated by undissolved dopant ions, which is not the case for the ion-implanted films. Increasing Ti valences are attributed to the fully oxidised condition of the Ti4+ ions that diffuse to the surface from Ti vacancy formation compared to the Ti valence of the bulk lattice, which contains Ti3+. Increasing O valence is attributed to the electronegativity of O2−, which is higher than that of Ti4+. Detailed structural mechanisms for the solubility and energetics mechanisms involve the initial formation of Mo and Cr interstitials that fill the two voids adjacent to the central Ti ion in the TiO6 octahedron, followed by integrated solid solubility (ISS) and intervalence/multivalence charge transfer (IVCT/MVCT). The sequential order of the last two is reversed for the two different doping methods. These two effects are likely to be the source of synergy, if any, between the two dopant ions. The photocatalytic performances of the uniformly codoped films are relatively poor and correlate well with the band gap (Eg). The performances of the ion-implanted films do not correlate with the Eg, where TiO2–Mo performs poorly but TiO2–Cr and TiO2–Mo–Cr outperform the undoped film. These results are interpreted in terms of the competition between the effects of Mo doping, which causes partial amorphisation and/or blockage of active sites, and Cr doping, which may cause Mo–Cr synergism, Cr-based heterojunction formation, and/or improved charge-carrier separation owing to the surface-deposition nature of ion implantation.