• We investigate the electrostatic effects in doped topological insulators by developing a self consistent scheme for an interacting tight binding model. The presence of bulk carriers, in addition to surface electrons, generates an intrinsic inhomogeneous charge density in the vicinity of the surface and, as a result, band bending effects are present. We find that electron doping and hole doping produce band bending effects of similar magnitude and opposite signs. The presence of additional surface dopants breaks this approximate electron-hole symmetry and dramatically affects the magnitude of the band bending. Applying a gate potential can generate a depletion zone characterized by a vanishing carrier density. We find that the density profile in the transition zone between the depleted region and the bulk is independent of the applied potential. In thin films the electrostatic effects are strongly dependent on the carrier charge density. In addition, we find that substrate induced potentials can generate a Rashba type spin-orbit coupling in ultra thin topological insulator films. We calculate the profiles of bulk and surface states in topological insulator films and identify the conditions corresponding to both types of states being localized within the same region in space.
  • We use the dynamical cluster approximation to understand the proximity of the superconducting dome to the quantum critical point in the two-dimensional Hubbard model. In a BCS formalism, $T_c$ may be enhanced through an increase in the d-wave pairing interaction ($V_d$) or the bare pairing susceptibility ($\chi_{0d}$). At optimal doping, where $V_d$ is revealed to be featureless, we find a power-law behavior of $\chi_{0d}(\omega=0)$, replacing the BCS log, and strongly enhanced $T_c$. We suggest experiments to verify our predictions.
  • High-temperature copper oxide superconductors (cuprates) display unconventional physics when they are lightly doped whereas the standard theory of metals prevails in the opposite regime. For example, the thermoelectric power, that is the voltage that develops across a sample in response to a temperature gradient, changes sign abruptly near optimal doping in a wide class of cuprates, a stark departure from the standard theory of metals in which the thermopower vanishes only when one electron exists per site. We show that this effect arises from proximity to a state in which particle-hole symmetry is dynamically generated. The operative mechanism is dynamical spectral weight transfer from states that lie at least 2eV away from the chemical potential. We show that the sign change is reproduced quantitatively within the Hubbard model for moderate values of the on-site repulsion, $U$. For sufficiently large values of on-site repulsion, for example, $U=20t$, ($t$ the hopping matrix element), dynamical spectral weight transfer attenuates and our calculated results for the thermopower are in prefect agreement with exact atomic limit. The emergent particle-hole symmetry close to optimal doping points to pairing in the cuprates being driven by high-energy electronic states.
  • We study the basic features of the local density of states (LDOS) observed in STM experiments on high-T$_c$ d-wave superconductors in the context of a minimal model of a d-wave superconductor which has {\it weakly} modulated off-diagonal disorder. We show that the low and high energy features of the LDOS are consistent with the observed experimental patterns and in particular, the anisotropic local domain features at high energies. At low energies, we obtain not only the scattering peaks predicted by the octet model, but also weak features that should be experimentally accessible. Finally, we show that the emerging features of the LDOS lose their correspondence with the features of the imposed disorder, as its complexity increases spatially.
  • This a comment on arXiv:0907.3243v2. We demonstrate that the method proposed by Efetov {\it et. al.} is just a reformulation of the Blankenbeckler, Scalapino, and Sugar approach and thus it contains exactly the same sign problem, including the dependence of the sign on the smoothness of the paths.
  • We compute the one-particle spectral function and the optical conductivity for the 2-d Hubbard model on a square lattice. The computational method is cellular dynamical mean-field theory (CDMFT) in which a 4-site Hubbard plaquette is embedded in a self-consistent bath. We obtain a `kink' feature in the dispersion of the spectral function and a mid-infrared (mid-IR) absorption peak in the optical conductivity, consistent with experimental data. Of the 256 plaquette states, only a single state which has d$_{x^2-y^2}$ symmetry contributes to the mid-IR, thereby suggesting a direct link with the pseudogap. Local correlations between doubly and singly occupied sites which lower the kinetic energy of a hole are the efficient cause of this effect.
  • We study the physics on the paramagnetic side of the phase diagram of the cobaltates, $Na_{x}CoO_{2}$, with an implementation of cellular dynamical mean field theory (CDMFT) with the non-crossing approximation (NCA) for the one-band Hubbard model on a triangular lattice. At low doping we find that the low energy physics is dominated by a quasi-dispersionless band. At half-filling, we find a metal-insulator transition at $U_{c}=5.6\pm0.15t$ which depends weakly on the cluster size. The onset of the metallic state occurs through the growth of a coherence peak at the chemical potential. Away from half filling, in the electron-doped regime, the system is metallic with a large, continuous Fermi surface as seen experimentally. Upon hole doping, a quasi non-dispersing band emerges at the top of the lower Hubbard band and controls the low-energy physics. This band is a clear signature of non-Fermi liquid behavior and cannot be captured by any weakly coupled approach. This quasi non-dispersive band, which persists in a certain range of dopings, has been observed experimentally. We also investigate the pseudogap phenomenon in the context of a triangular lattice and we propose a new framework for discussing the pseudogap phenomena in general. This framework involves a momentum-dependent characterization of the low-energy physics and links the appearance of the pseudogap to a reconstruction of the Fermi surface without invoking any long range order or symmetry breaking. Within this framework we predict the existence of a pseudogap for the two dimensional Hubbard model on a triangular lattice in the weakly hole-doped regime.
  • We demonstrate that at a filling of $n=1.5$, an hexatic insulating state obtains in the extended Hubbard model on a triangular lattice. Composed of two tetragonal sublattices with fillings of $n=1$ and $n=2$, the insulating state is charge ordered and possesses an antiferromagnetic superlattice with dimension $a\times\sqrt{3}$. Two distinct energy scales arise in our model, a charge gap for the insulator and the effective exchange interaction in the antiferromagnet. Our model is capable of explaining both qualitatively and quantitatively the Hall coefficient including the sign change, the temperature dependence of the resistivity and the persistence of antiferromagnetism above the insulating state.
  • We show that doped Mott insulators, such as the copper-oxide superconductors, are asymptotically slaved in that the quasiparticle weight, $Z$, near half-filling depends critically on the existence of the high energy scale set by the upper Hubbard band. In particular, near half filling, the following dichotomy arises: $Z\ne 0$ when the high energy scale is integrated out but Z=0 in the thermodynamic limit when it is retained. Slavery to the high energy scale arises from quantum interference between electronic excitations across the Mott gap. Broad spectral features seen in photoemission in the normal state of the cuprates are argued to arise from high energy slavery.
  • Three well-known perturbative approaches to deriving low-energy effective theories, the degenerate Brillouin-Wigner perturbation theory (projection method), the canonical transformation, and the resolvent methods are compared. We use the Hubbard model as an example to show how, to fourth order in hopping t, all methods lead to the same effective theory, namely the t-J model with ring exchange and various correlated hoppings. We emphasize subtle technical difficulties that make such a derivation less trivial to carry out for orders higher than second. We also show that in higher orders, different approaches can lead to seemingly different forms for the low-energy Hamiltonian. All of these forms are equivalent since they are connected by an additional unitary transformation whose generator is given explicitly. The importance of transforming the operators is emphasized and the equivalence of their transformed structure within the different approaches is also demonstrated.
  • This paper has been withdrawn by the authors.