Polaronic effects in TiO<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mrow><mml:msub><mml:mrow /><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub></mml:mrow></mml:math>calculated by the HSE06 hybrid functional: Dopant passivation by carrier self-trapping

Deák, Péter; Aradi, Bálint; Frauenheim, Thomas

doi:10.1103/physrevb.83.155207

Cited by 189 publications

(135 citation statements)

References 70 publications

Supporting

Mentioning

113

Contrasting

Order By: Relevance

“…HSE06 can overestimate band gaps of metal oxides, e.g. for TiO 2 [44] or d 6 spinels [45], so this finding is not so unusual.. This finding is consistent with the now comprehensive literature in which comparisons of the band gaps of solids from DFT, DFT+U and hybrid DFT are available.…”

Section: Undoped Bulk Niosupporting

confidence: 80%

Hybrid density functional theory description of N- and C-doping of NiO

Nolan

Long

English

et al. 2011

The Journal of Chemical Physics

View full text Add to dashboard Cite

The large intrinsic bandgap of NiO hinders its potential application as a photocatalyst under visible-light irradiation. In this study, we have performed first-principles screened exchange hybrid density functional theory with the HSE06 functional calculations of N- and C-doped NiO to investigate the effect of doping on the electronic structure of NiO. C-doping at an oxygen site induces gap states due to the dopant, the positions of which suggest that the top of the valence band is made up primarily of C 2p-derived states with some Ni 3d contributions, and the lowest-energy empty state is in the middle of the gap. This leads to an effective bandgap of 1.7 eV, which is of potential interest for photocatalytic applications. N-doping induces comparatively little dopant-Ni 3d interactions, but results in similar positions of dopant-induced states, i.e., the top of the valence band is made up of dopant 2p states and the lowest unoccupied state is the empty gap state derived from the dopant, leading to bandgap narrowing. With the hybrid density functional theory (DFT) results available, we discuss issues with the DFT corrected for on-site Coulomb description of these systems

show abstract

Section: Undoped Bulk Niosupporting

confidence: 80%

Hybrid density functional theory description of N- and C-doping of NiO

Nolan

Long

English

et al. 2011

The Journal of Chemical Physics

View full text Add to dashboard Cite

show abstract

“…Janotti et al [37] and Deak [38] have employed the widely used hybrid function HSE06 functional [39,40] to study the electronic properties of the oxygen vacancy and substitutional dopants. They found that HSE gave a greatly improved description than GGA, but they did not treat the Ti interstitial.…”

Section: Introductionmentioning

confidence: 99%

Calculation of point defects in rutile TiO2by the screened-exchange hybrid functional

2012

View full text Add to dashboard Cite

The formation energies of the oxygen vacancy and titanium interstitial in rutile TiO 2 were calculated by the screened exchange (sX) hybrid density functional method, which gives a band gap of 3.1 eV, close to the experimental value. The O vacancy gives rise to a gap state lying 0.7 eV below the conduction band edge, whose charge density is localised around the two of three Ti atoms next to the vacancy. The Ti interstitial generates four defect states in the gap, whose unpaired electrons lie on the interstitial and the adjacent Ti 3d orbitals. The formation energy for the neutral O vacancy is 1.9 eV for the O-poor chemical potential, and similar to that of the neutral Ti interstitial, and has a lower formation energy for Ti interstitial under O-rich conditions. This indicates that both the O vacancy and Ti interstitial are relevant for oxygen deficiency in rutile TiO 2 but the O vacancy will dominate under O-rich conditions. This resolves the questions about defect localisation and defect predominance in the literature.

show abstract

“…However, the sharpness of the transition between ascending and descending slopes seen for log[A(ω H )/A(ω L )] vs T -1 in Fig. 4 [28][29][30] in which either correlation or hybridization was included in the density functional formalism. We use these and other results to address the origin of the energy levels that emerge from analysis of the experimental results.…”

mentioning

confidence: 99%