All-electron self-consistent GW approximation: application to Si, MnO, and NiO

Quasiparticle self-consistent GW method: a short summary

We have developed a quasiparticle self-consistent GW method (QSGW), which is a new self-consistent method to calculate the electronic structure within the GW approximation. The method is formulated based on the idea of a self-consistent perturbation; the non-interacting Green function G(0), which is the starting point for GWA to obtain G, is determined self-consistently so as to minimize the perturbative correction generated by GWA. After self-consistency is attained, we have G(0), W (the screened Coulomb interaction) and G self-consistently. This G(0) can be interpreted as the optimum non-interacting propagator for the quasiparticles. We will summarize some theoretical discussions to justify QSGW. Then we will survey results which have been obtained up to now: e.g., band gaps for normal semiconductors are predicted to a precision of 0.1-0.3 eV; the self-consistency including the off-diagonal part is required for NiO and MnO; and so on. There are still some remaining disagreements with experiments; however, they are very systematic, and can be explained from the neglect of excitonic effects.

Effect of surrounding point charges on the density functional calculations of NixOx clusters (x = 4–12)

Embedded Ni(x)O(x) clusters (x = 4-12) have been studied by the density-functional method using compensating point charges of variable magnitude to calculate the ionic charge, bulk modulus, and lattice binding energy. The computations were found to be strongly dependent on the value of the surrounding point charge array and an optimum value could be found by choosing the point charge to reproduce the experimentally observed Ni--O lattice parameter. This simple, empirical method yields a good match between computed and experimental data, and even small variation from the optimum point charge value produces significant deviation between computed and measured bulk physical parameters. The optimum point charge value depends on the cluster size, but in all cases is significantly less than +/-2.0, the formal oxidation state typically employed in cluster modeling of NiO bulk and surface properties. The electronic structure calculated with the optimized point charge magnitude is in general agreement with literature photoemission and XPS data and agrees with the presently accepted picture of the valence band as containing charge-transfer insulator characteristics. The orbital population near the Fermi level does not depend on the cluster size and is characterized by hybridized Ni 3d and O 2p orbitals with relative oxygen contribution of about 70%.

Exchange and correlation effects in electronic excitations of Cu(2)O

State-of-the-art theoretical methods fail in describing the optical absorption spectrum, band gap, and optical onset of Cu(2)O. We have extended a recently proposed self-consistent quasiparticle approach, based on the GW approximation, to the calculation of optical spectra, including excitonic effects. The band structure compares favorably with our present angle-resolved photoemission measurements. The excitonic effects based on these realistic band structure and screening provide a reliable optical absorption spectrum, which allows for a revised interpretation of its main structures.

Local self-energy approach for electronic structure calculations

Using a novel self-consistent implementation of Hedin's perturbation theory, we calculate space- and energy-dependent self-energy for a number of materials. We find it to be local in real space and rapidly convergent on second- to third-nearest neighbors. Corrections beyond are evaluated and shown to be completely localized within a single unit cell. This can be viewed as a fully self-consistent implementation of the dynamical mean field theory for electronic structure calculations of real solids using a perturbative impurity solver.

Quasiparticle self-consistent GW theory

In past decades the scientific community has been looking for a reliable first-principles method to predict the electronic structure of solids with high accuracy. Here we present an approach which we call the quasiparticle self-consistent approximation. It is based on a kind of self-consistent perturbation theory, where the self-consistency is constructed to minimize the perturbation. We apply it to selections from different classes of materials, including alkali metals, semiconductors, wide band gap insulators, transition metals, transition metal oxides, magnetic insulators, and rare earth compounds. Apart from some mild exceptions, the properties are very well described, particularly in weakly correlated cases. Self-consistency dramatically improves agreement with experiment, and is sometimes essential. Discrepancies with experiment are systematic, and can be explained in terms of approximations made.

Ab initio prediction of conduction band spin splitting in zinc blende semiconductors

We use a recently developed self-consistent GW approximation to present systematic ab initio calculations of the conduction band spin splitting in III-V and II-VI zinc blende semiconductors. The spin-orbit interaction is taken into account as a perturbation to the scalar relativistic Hamiltonian. These are the first calculations of conduction band spin splittings based on a quasiparticle approach; and because the self-consistent GW scheme accurately reproduces the relevant band parameters, it is expected to be a reliable predictor of spin splittings. The results are compared to the few available experimental data and a previous calculation based on a model one-particle potential. We also briefly address the widely used k x p parametrization in the context of these results.

Ab initio treatment of optical second harmonic generation in NiO

In this Letter we develop a new systematic approach to study optical second harmonic generation in NiO, on both the (001) surface and the bulk. NiO is modeled as a doubly embedded cluster on which two highly correlated quantum chemistry methods are applied in order to obtain the wave functions of all the intragap d states and the low lying charge transfer states. The optical gap is calculated and the electric dipole, magnetic dipole, and electric quadrupole contributions to the second order susceptibility tensor are computed for the first time from first principles. Going beyond the electric dipole approximation gives new insight into the experimentally observed spectrum. A method is proposed for monitoring the spin dynamics of the NiO(001) surface.