No abstract

Self-consistent GW calculations, maintaining only the quasiparticle part of the Green's function G, are reported for a wide class of materials, including small gap semiconductors and large gap insulators. We show that the inclusion of the attractive electron-hole interaction via an effective nonlocal exchange correlation kernel is required to obtain accurate band gaps in the framework of self-consistent GW calculations. If these are accounted for via vertex corrections in W, the band gaps are found to be within a few percent of the experimental values.

Algorithmic details and results of fully frequency-dependent G 0 W 0 calculations are presented. The implementation relies on the spectral representation of the involved matrices and their Hilbert or Kramers-Kronig transforms to obtain the polarizability and self-energy matrices at each frequency. Using this approach, the computational time for the calculation of polarizability matrices and quasiparticle energies is twice as that for a single frequency, plus Hilbert transforms. In addition, the implementation relies on the PAW method, which allows to treat d-states with relatively modest effort and permits the reevaluation of the core-valence interaction on the level of the Hartree-Fock approximation. Tests performed on an sp material ͑Si͒ and materials with d electrons ͑GaAs and CdS͒ yield quasiparticle energies that are very close to previous all-electron pseudopotential and all-electron full-potential linear muffin-tin-orbital calculations.

We present a comparative full-potential study of generalized Kohn-Sham schemes (gKS) with explicit focus on their suitability as starting point for the solution of the quasiparticle equation. We compare $G_0W_0$ quasiparticle band structures calculated upon LDA, sX, HSE03, PBE0, and HF functionals for exchange and correlation (XC) for Si, InN and ZnO. Furthermore, the HSE03 functional is studied and compared to the GGA for 15 non-metallic materials for its use as a starting point in the calculation of quasiparticle excitation energies. For this case, also the effects of selfconsistency in the $GW$ self-energy are analysed. It is shown that the use of a gKS scheme as a starting point for a perturbative QP correction can improve upon the deficiencies found for LDA or GGA staring points for compounds with shallow $d$ bands. For these solids, the order of the valence and conduction bands is often inverted using local or semi-local approximations for XC, which makes perturbative $G_0W_0$ calculations unreliable. The use of a gKS starting point allows for the calculation of fairly accurate band gaps even in these difficult cases, and generally single-shot $G_0W_0$ calculations following calculations using the HSE03 functional are very close to experiment

The well-ordered aluminum oxide film formed by oxidation of the NiAl(110) surface is the most intensely studied metal surface oxide, but its structure was previously unknown. We determined the structure by extensive ab initio modeling and scanning tunneling microscopy experiments. Because the topmost aluminum atoms are pyramidally and tetrahedrally coordinated, the surface is different from all Al2O3 bulk phases. The film is a wide-gap insulator, although the overall stoichiometry of the film is not Al2O3 but Al10O13. We propose that the same building blocks can be found on the surfaces of bulk oxides, such as the reduced corundum (0001) surface.

Activation of fuel molecules (H2, CH4, CO) at the anode triple phase boundary (TPB) of a solid oxide fuel cell, modeled by a Ni (nickel)/YSZ (yttria-stabilized zirconia)/fuel interface is investigated using density functional theory. We demonstrate that, by employing ab initio calculations, it is possible to elucidate the mechanisms of electronic charge transfer and current generation as a result of electrochemical oxidation of fuel in the anode TPB. Moreover, we show that an oxygen-enriched YSZ surface (YSZ+O) of the Ni/YSZ cermet is significantly less active toward oxidation of fuel molecules than an oxygen-enriched YSZ surface in the absence of Ni, which is explained by partial saturation of the valence of an extra oxygen atom of the Ni/YSZ cermet. On the other hand, the chemical activity of the Ni part of the cermet is close to that of the infinite Ni surface, even in the proximity to the YSZ. The YSZ+O surface of the cermet is found generally inert, as adsorption of H2 and CH4 is associated with high kinetic barriers, whereas CO adsorption is a more favorable, although endothermic reaction. The oxidation of H2 and CO on Ni after oxygen spillover from the oxide is exothermic, and CH4 oxidation is mildly endothermic. It is also shown that transfer of hydrogen atoms from Ni to YSZ (hydrogen spillover reactions) with subsequent water formation is another possible scenario of hydrogen oxidation on the YSZ+O surface.

A model for the straight antiphase domain boundary of the ultrathin aluminum oxide film on the NiAl(110) substrate is derived from scanning tunneling microscopy measurements and density-functional theory calculations. Although the local bonding environment of the perfect film is maintained, the structure is oxygen deficient and possesses a favorable adsorption site. The domain boundary exhibits a downwards band bending and three characteristic unoccupied electronic states, in excellent agreement with scanning tunneling spectroscopy measurements.

A new mechanism for hydrogen oxidation on the nickel/yttria-stabilized zirconia (Ni/YSZ) interface is proposed based on density functional theory. The new mechanism involves oxidation of hydrogen by the oxygen atoms that are bound to both nickel and zirconium (or yttrium) at the interface. The free energy change (∆G) for this pathway is compared to ∆G for reaction steps of previously proposed oxidation mechanisms involving spillover of oxygen from YSZ to nickel or hydrogen spillover from nickel to YSZ. For all mechanisms, we consider both a stoichiometric (YSZ) surface as well as an oxygen enriched YSZ surface (YSZ+O) where in the latter case a vacant site is filled by an oxygen atom transferred from the YSZ bulk. The release of water as the final product in hydrogen oxidation is facilitated at high temperatures by entropy. The difference between the current and previous mechanisms is that for the hydrogen oxidation now we only consider the involvement of oxygen atoms that are bound to both nickel and zirconium (or yttrium). In previous studies we only considered oxygen atoms that initially were bound to zirconium (or yttrium) only.