V. S. Oudovenko

Thermoelectric properties of the degenerate Hubbard model

We investigate the thermoelectric properties of a system near a pressure-driven Mott-Hubbard transition. The dependence of the thermopower and the figure of merit on pressure and temperature within a degenerate Hubbard model for integer filling n = I is calculated using dynamical mean-field theory. The quantum Monte Carlo method is used to solve the impurity model. Obtained results can qualitatively explain thermoelectric properties of various strongly correlated materials.

On the theory of superconductivity in the extended Hubbard model

A microscopic theory of superconductivity in the extended Hubbard model which takes into account the intersite Coulomb repulsion and electron-phonon interaction is developed in the limit of strong correlations. The Dyson equation for normal and pair Green functions expressed in terms of the Hubbard operators is derived. The self-energy is obtained in the noncrossing approximation. In the normal state, antiferromagnetic short-range correlations result in the electronic spectrum with a narrow bandwidth. We calculate superconducting Tc by taking into account the pairing mediated by charge and spin fluctuations and phonons. We found the d-wave pairing with high-Tc mediated by spin fluctuations induced by the strong kinematic interaction for the Hubbard operators. Contributions to the d-wave pairing coming from the intersite Coulomb repulsion and phonons turned out to be small.

Electronic structure and magnetic anisotropy ofCrO2

The problem of importance of strong correlations for the electronic structure, transport and magnetic properties of half--metallic ferromagnetic CrO_2 is addressed by performing density functional electronic structure calculations in the local spin density approximation (LSDA) as well as using the LSDA+U method. It is shown that the corresponding low--temperature experimental data are best fitted without accounting for the Hubbard U corrections. We conclude that the ordered phase of CrO

Quantum Monte Carlo calculation of the finite temperature Mott-Hubbard transition

_2 is weakly correlated.

The α → γ transition in Ce: A theoretical view from optical spectroscopy

We present clear numerical evidence for the coexistence of metallic and insulating dynamical mean-field theory (DMFT) solutions in a half-filled single-band Hubbard model at finite temperature. The quantum Monte Carlo (QMC) method is used to solve the DMFT equation. We discuss important technical aspects of the DMFT-QMC method, which should be properly dealt with in order to obtain reliable DMFT solutions at low temperatures. Among these technical aspects are the critical slowing down near the boundaries of the coexistence region, the convergence criteria for the DMFT iterations, and the correct interpolation of the discretized Greens functions.

Kinematic spin-fluctuation mechanism of high-temperature superconductivity

Using a novel approach to calculate optical properties of strongly correlated systems, we address the old question of the physical origin of the alpha--> gamma transition in Ce. We find that the Kondo collapse model, involving both the f and the spd electrons, describes the optical data better than a Mott transition picture involving the f electrons only. Our results compare well with existing experiments on thin films. We predict the full temperature dependence of the optical spectra and find the development of a hybridization pseudogap in the vicinity of the alpha--> gamma phase transition.

Dynamical correlations in multiorbital Hubbard models: fluctuation exchange approximations

We study d-wave superconductivity in the extended Hubbard model in the strong correlation limit for a large intersite Coulomb repulsion V. We argue that in the Mott-Hubbard regime with two Hubbard subbands, there emerges a new energy scale for the spin-fluctuation coupling of electrons of the order of the electron kinetic energy W much larger than the exchange energy J. This coupling is induced by the kinematic interaction for the Hubbard operators, which results in the kinematic spin-fluctuation pairing mechanism for V ≲ W. The theory is based on the Mori projection technique in the equation of motion method for the Green’s functions in terms of the Hubbard operators. The doping dependence of the superconductivity temperature Tc is calculated for various values of U and V.

Interpolative approach for solving the Anderson impurity model

We study the two-band degenerate Hubbard model using the fluctuation exchange approximation (FLEX) and compare the results with quantum Monte Carlo (QMC) calculations. Both the self-consistent and the non-self-consistent versions of the FLEX scheme are investigated. We find that, unlike in the one-band case, in the multiband case, good agreement with the quantum Monte Carlo results is obtained within the electron-electron T-matrix approximation using the full renormalization of the one-particle propagators. The crossover to strong coupling and the formation of satellites is more clearly visible in the non-self-consistent scheme. Finally we discuss the behaviour of the FLEX for higher orbital degeneracy.

Linearized self-consistent quasiparticle GW method: Application to semiconductors and simple metals

A rational representation for the self-energy is explored to interpolate the solution of the Anderson impurity model in the general orbitally degenerate case. Several constraints such as Friedels sum rule and the positions of the Hubbard bands, as well as the value of the quasiparticle residue, are used to establish the equations for the coefficients of the interpolation. We employ two fast techniques, the slave-boson mean-field and the Hubbard I approximations, to determine the functional dependence of the coefficients on doping, degeneracy, and the strength of the interaction. The obtained spectral functions and self-energies are in good agreement with the results of the numerically exact quantum Monte Carlo method.

The alpha to gamma transition in Ce: a theoretical view from optical spectroscopy

Abstract We present a code implementing the linearized quasiparticle self-consistent GW method (LQSGW) in the LAPW basis. Our approach is based on the linearization of the self-energy around zero frequency which differs it from the existing implementations of the QSGW method. The linearization allows us to use Matsubara frequencies instead of working on the real axis. This results in efficiency gains by switching to the imaginary time representation in the same way as in the space time method. The all electron LAPW basis set eliminates the need for pseudopotentials. We discuss the advantages of our approach, such as its N 3 scaling with the system size N , as well as its shortcomings. We apply our approach to study the electronic properties of selected semiconductors, insulators, and simple metals and show that our code produces the results very close to the previously published QSGW data. Our implementation is a good platform for further many body diagrammatic resummations such as the vertex-corrected GW approach and the GW+DMFT method. Program summary Program Title: LqsgwFlapw Program Files doi: http://dx.doi.org/10.17632/cpchkfty4w.1 Licensing provisions: GNU General Public License Programming language: Fortran 90 External routines/libraries: BLAS, LAPACK, MPI (optional) Nature of problem: Direct implementation of the GW method scales as N 4 with the system size, which quickly becomes prohibitively time consuming even in the modern computers. Solution method: We implemented the GW approach using a method that switches between real space and momentum space representations. Some operations are faster in real space, whereas others are more computationally efficient in the reciprocal space. This makes our approach scale as N 3 . Restrictions: The limiting factor is usually the memory available in a computer. Using 10 GB/core of memory allows us to study the systems up to 15 atoms per unit cell.