Figure 4. (a) High-resolution microscopy image of the c-Si/SiO2/ZnTiO3/Al contact; (b) HAADF–STEM image, EDX mapping and line scan of 300 °C-annealed c-Si/SiO2/ZnTiO3/Al contact; (c, d) comparison of HAADF–STEM microscopy images of an as-fabricated sample and a 300 °C-annealed sample.
To identify the valence state of Al occurring in the ZnTiO3 film, high-resolution XPS characterizations have been performed on the ZnTiO3 film with and without annealing. Note that the ZnTiO3 film is also capped by an Al layer, just like in solar cells, but the Al layer is removed by a wet chemical method prior to the XPS measurement. More details can be found in experimental part. The spectra of Al 2p core levels are presented in Figure 5a. For the film without annealing, two Al peaks can be obtained. The strong peak at the binding energy of 74.4 eV corresponds to the Al-O bond, which slightly shifts towards lower binding energies compared to the 74.6 eV peak position of Al 2p3/2 in stoichiometric Al2O3,46,47 indicating that Al occurred in the ZnTiO3 film plays a role as dopant (ZnTiO3:Al). The peak centered at 72.3 eV is a characteristic of metallic Al.48 For the sample annealed at 300 ℃, no metallic Al peak is observed, confirming that post-annealing further promotes the doping of Al in ZnTiO3 films, and made Al-O bonds in zinc titanate completely. The doping of Al changes the WF of the ZnTiO3 film, which is verified by UPS. As shown in Figure 5b, a WF value of 3.1 eV can be extracted for the as-fabricated Al-ZnTiO3film, which is lower compared to the pure ZnTiO3 film (4.0 eV).40 After 300 ℃ post-annealing, the WFof Al-ZnTiO3 film further decreases to 2.07 eV.