arXiv subjects
Publications and source records attributed to Gabriel Kotliar.
The extended Dynamical Mean Field Equations (EDMFT) are analyzed using semiclassical methods for a model describing an interacting fermi-bose system. We compare the semiclassical approach with the exact QMC (Quantum Montecarlo) method. We found the transition to an ordered state to be of the first order for any dimension below four.
We develop the extended dynamical mean field theory (E-DMFT) with a view towards realistic applications. {\bf 1)} We introduce an intuitive derivation of the E-DMFT formalism. By identifying the Hartree contributions before the E-DMFT treatment, it allows to handle systems in symmetry breaking phases within a simple formalism. {\bf 2)} We make a new implementation of E-DMFT through real Hubbard-Stratonovich transformation to decouple the non-local two-particle interactions. We apply it to a 3D U-V model and investigate the behavior of the various Green's functions, especially the density susceptibility, as the density instability is approached. We obtain the phase diagram at a finite temperature. {\bf 3)} We present a formalism incorporating E-DMFT with Cellular DMFT. {\bf 4)} We suggest an improvement of the E-DMFT approach by combining it with a generalized GW method. The method combines the local self-energy from E-DMFT and the non-local ones from the perturbative calculation of GW. We apply the method to a 1D U-V model with two sublattices carrying different chemical potentials. By comparing with those from DMRG, we show the results are shifted in the correct direction due to the GW contributions. {\bf 5)} In order to handle the generic Coulomb repulsion within E-DMFT, we describe a method to tailor E-DMFT so that proper momentum dependence can be kept in general response functions.
We develop, clarify and test various aspects of cluster methods dynamical mean field methods using a soluble toy model as a benchmark. We find that the Cellular Dynamical Mean Field Theory (C-DMFT) converges very rapidly and compare its convergence properties with those of the Dynamical Cluster Approximation (DCA). We propose and test improved estimators for the lattice self energy within C-DMFT.
We calculate magnetic anisotropy energy of Fe and Ni by taking into account the effects of strong electronic correlations, spin-orbit coupling, and non-collinearity of intra--atomic magnetization. The LDA+U method is used and its equivalence to dynamical mean--field theory in the static limit is emphasized. Both experimental magnitude of MAE and direction of magnetization are predicted correctly near U=1.9 eV, J=1.2 eV for Ni and U=1.2 eV, J=0.8 eV for Fe. Correlations modify the one--electron spectra which are now in better agreement with experiments.
We propose a cellular version of dynamical-mean field theory which gives a natural generalization of its original single-site construction and is formulated in different sets of variables. We show how non-orthogonality of the tight-binding basis sets enters the problem and prove that the resulting equations lead to manifestly causal self energies.
We calculate magnetic anisotropy energy of Fe and Ni by taking into account the effects of strong electronic correlations, spin-orbit coupling, and non-collinearity of intra-atomic magnetization. The LDA+U method is used and its equivalence to dynamical mean-field theory in the static limit is emphasized. Both experimental magnitude of MAE and direction of magnetization are predicted correctly near U=4 eV for Ni and U=3.5 eV for Fe. Correlations modify one-electron spectra which are now in better agreement with experiments.
We study the nonlinear feedback in a fermion-boson system using an extension of dynamical mean-field theory and the quantum Monte Carlo method. In the perturbative regimes (weak-coupling and atomic limits) the effective interaction among fermions increases as the width of the boson dispersion increases. In the strong coupling regime away from the anti-adiabatic limit, the effective interaction decreases as we increase the width of the boson dispersion. This behavior is closely related with complete softening of the boson field. We elucidate the parameters that control this nonperturbative region where fluctuations of the dispersive bosons enhance the delocalization of fermions.
We investigate the paramagnetic-metal-to-antiferromagnetic-metal and antiferromagnetic-metal-to- antiferromagnetic-insulator transitions using a slave-boson mean-field theory. To this effect, we discuss the ground state of the half-filled Hubbard model as a function of t'/t and correlation strength U, where t and t' are the hopping amplitudes between nearest and next-nearest neighbors, respectively. The metal-insulator transition at a critical U_{MIT} is of second order for small levels of magnetic frustration, t'/t<0.06, and of first order for large ones, t'/t>0.06. The insulator is always antiferromagnetically ordered, while the metal exhibits a second-order transition from a paramagnetic to an antiferromagnetic state up to t'/t=0.14, as U is increased. We also contrast these findings with what we obtain in Hartree-Fock approximation.
We investigate the thermoelectric response of correlated electron systems near the density driven Mott transition using the dynamical mean field theory.
We study the interactions between the coherent quasiparticles and the incoherent Mott-Hubbard excitations and their effects on the low energy properties in the $U=\infty$ Hubbard model. Within the framework of a systematic large-N expansion, these effects first occur in the next to leading order in 1/N. We calculate the scattering phase shift and the free energy, and determine the quasiparticle weight Z, mass renormalization, and the compressibility. It is found that the compressibility is strongly renormalized and diverges at a critical doping $δ_c=0.07\pm0.01$. We discuss the nature of this zero-temperature phase transition and its connection to phase separation and superconductivity.
We discuss the solution of the Mott transition problem in a fully frustrated lattice with a semicircular density of states in the limit of infinite dimensions from the point of view of a Landau free energy functional. This approach provides a simple relation between the free energy of the lattice model and that of its local description in terms of an impurity model. The character of the Mott transition in infinite dimensions, (as reviewed by Georges Kotliar Krauth and Rozenberg, RMP 68, 1996, 13) follows simply from the form of the free energy functional and the physics of quantum impurity models. At zero temperature, below a critical value of the interaction U, a Mott insulator with a finite gap in the one particle spectrum, becomes unstable to the formation of a narrow band near the Fermi energy. Using the insights provided by the Landau approach we answer questions raised about the dynamical mean field solution of the Mott transition problem, and comment on its applicability to three dimensional transition metal oxides.
We derive new sum rules for the real and imaginary parts of the frequency-dependent Hall constant and Hall conductivity. As an example, we discuss their relevance to the doped Mott insulator that we describe within the dynamical mean-field theory of strongly correlated electron systems.
We investigate the Hall effect and the magnetoresistance of strongly correlated electron systems using the dynamical mean-field theory. We treat the low- and high-temperature limits analytically and explore some aspects of the intermediate-temperature regime numerically. We observe that a bipartite-lattice condition is responsible for the high-temperature result $σ_{xy}\sim 1/T^2$ obtained by various authors, whereas the general behavior is $σ_{xy}\sim 1/T$, as for the longitudinal conductivity. We find that Kohler's rule is neither obeyed at high nor at intermediate temperatures.
The multichannel Kondo model with SU(N) spin symmetry and SU(K) channel symmetry is considered. The impurity spin is chosen to transform as an antisymmetric representation of SU(N), corresponding to a fixed number of Abrikosov fermions $\sum_αf_α^{\dagger}f_α=Q$. For more than one channel (K>1), and all values of N and Q, the model displays non-Fermi behaviour associated with the overscreening of the impurity spin. Universal low-temperature thermodynamic and transport properties of this non-Fermi liquid state are computed using conformal field theory methods. A large-N limit of the model is then considered, in which K/N and Q/N are held fixed. Spectral densities satisfy coupled integral equations in this limit, corresponding to a (time-dependent) saddle-point. A low frequency, low-temperature analysis of these equations reveals universal scaling properties in the variable $ω/T$, also predicted from conformal invariance. The universal scaling form is obtained analytically and used to compute the low-temperature universal properties of the model in the large-N limit, such as the T=0 residual entropy and residual resistivity, and the critical exponents associated with the specific heat and susceptibility. The connections with the ``non-crossing approximation'' and the previous work of Cox and Ruckenstein are discussed.
A slave boson representation for the degenerate Hubbard model is introduced. The location of the metal to insulator transition that occurs at commensurate densities is shown to depend weakly on the band degeneracy M. The relative weights of the Hubbard sub-bands depend strongly on M, as well as the magnetic properties. It is also shown that a sizable Hund's rule coupling is required in order to have a ferromagnetic instability appearing. The metal to insulator transition driven by an increase in temperature is a strong function of it.
We investigate the Mott transition in infinite dimensions in the orbitally degenerate Hubbard model. We find that the qualitative features of the Mott transition found in the one band model are also present in the orbitally degenerate case. Surprisingly, the quantitative aspects of the transition around density one are not very sensitive to orbital degeneracy, justifying the quantitative success of the one band model which was previously applied to orbitally degenerate systems. We contrast this with quantities that have a sizeable dependence on the orbital degeneracy and comment on the role of the intraatomic exchange J.
We interpret the upper-critical-field anomalies observed in some high- temperature superconductors as resulting from the proximity to a zero-temperature quantum critical point. We estimate the shape of the phase boundary between the normal and the superconducting phase by modeling the zero temperature critical point as the second-order endpoint of the first-order melting line of the vortex lattice.
We study the mixed valence regime of a generalized Anderson impurity model using the bosonization approach. This single impurity problem is defined by the $U=\infty$ Anderson model with an additional density-density interaction, as well as an explicit exchange interaction, between the impurity and conduction electrons. We find three points in the interaction parameter space at which all the correlation functions can be calculated explicitly. These points represent the mixed valence counterparts of the ususal Toulouse point for the Kondo problem, and are appropriately named the Toulouse points of the mixed valence problem. Two of the Toulouse points exhibit the strong coupling, Fermi liquid behavior. The third one shows spin-charge separation; here, the spin-spin correlation functions are Fermi-liquid-like, the charge-charge correlation functions and the single particle Green function have non-Fermi-liquid behaviors, and a pairing correlation function is enhanced compared to the Fermi liquid case. This third Toulouse point describes the novel intermediate mixed valence phase we have previously identified. In deriving these results, we emphasize the importance of keeping track of the anticommutation relation between the fermion fields when the bosonization method is applied to quantum impurity problems.