Photoemission and Bremsstrahlung Isochromat Spectroscopy Studies of TiO₂(Rutile) and SrTiO₃

Self-energy and excitonic effects in the electronic and optical properties ofTiO2crystalline phases

“…The optical gaps ͑3.0-3.2 eV͒ are larger than the DFT ones and always smaller than the electronic ones calculated with many-body corrections, due to excitonic effects, phonon-assisted indirect transitions, and temperature effects. When the calculated electronic gap at many-body level is properly compared 20 to photoemission results, 50,53,63 as it is done in the present paper, a quite good agreement can be found. Some calculations of the optical spectra of these polymorphs 22,30,34,56,59 have been performed at the randomphase approximation ͑RPA͒ level, but this description is inappropriate to correctly take into account the optical response of the material, as we show in the results section.…”

“…We do not report here these results because they are in line with data already published. 20,23,30,50,97 We can say that the comparison with experimental photoemission spectra, 42,44,47,50,53,63 integrated or angular resolved, is quite good, in spite of the well-known difficulties of DFT in describing the d-electron properties. 16 To take into account the importance of a different wave function starting point for the G 0 W 0 calculations onto the electronic description of bulk TiO 2 , we compare the spatial localization of Kohn-Sham wave functions obtained from DFT-PBE ͓Ref.…”

“…32-45͒ and have been reinvestigated more recently. [46][47][48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65] From the photoemission data on rutile 42,44,47,50,53,63 the electronic band gap, corresponding to the difference between the valence-band maximum ͑VBM͒ and the conduction-band minimum ͑CBM͒, has reported values of 3.3Ϯ 0.5 eV ͑Ref. 53͒ ͓ultraviolet photoemission spectroscopy ͑UPS͒ and inverse photoemission spectroscopy ͑IPES͔͒, 3.6Ϯ 0.2 eV ͑Ref.…”

Optical properties of tungsten trioxide from first-principles calculations

“…The HSE06 functional [44,45] and the ∆SCF method yield very similar results for both γ-WO 3 and simple cubic WO 3 , and they appear to moderately underestimate photoemission experiments; analogous findings were reported for rutile TiO 2 , whose gap computed with the HSE06 functional, 3.05 eV [51], underestimates the measured photoemission gap (3.3-3.6 eV [19,20] However additional, important effects need to be taken into account, before carrying out a meaningful comparison with experiment, e.g. spin-orbit (SO) effects and corrections to the computed gap due to electron phonon interaction.…”

“…Several measurements by UV-vis spectroscopy [16] and photoelectrolysis [17] yielded an indirect optical gap of 2.6 eV at room temperature(T), while Salje et al [18], who measured transmission spectra at room temperature, reported a direct gap of 2.58 eV. Similar to the case of TiO 2 [7,19,20], direct and inverse photoemission measurements of the fundamental gap of WO 3 led to a value much larger (0.6∼0.7 eV) [21,22] than that of its optical gap, and this difference cannot be accounted for by the exciton binding energy. We note that optical and photoemission experiments were both conducted on the phase stable at room T. On the theoretical side, a coherent and consistent interpretation of experiments has not yet been formulated and the level of theory necessary to describe photoemission and absorption experiments of WO 3 is yet unclear.…”