arXiv subjects
Publications and source records attributed to Yu-Zhong Zhang.
Using the nonlocal coherent-potential approximation we study the effect of intersite spatial correlations on the transition from band insulator to metal as well as from metal to Mott insulator in the "alloy analogy" approach to the paramagnetic solution of the half-filled ionic Hubbard model. We find that intersite spatial correlations enhance the metallic phase.
The origin of magnetism and superconductivity in iron-based superconductors is still unclear. Here, by investigating the momentum-dependent particle-hole excitations which quantify the tendency of itinerant electrons towards various magnetic states or superconducting phases, we unravel a novel origin to account for the variety of physical properties of iron-based compounds. We show that condensation of particle-hole excitations away from the Fermi surface in momentum space is the underlying mechanism in deciding the magnetic and superconducting properties of iron-based materials. The applicability of this scenario to the whole family of iron-based superconductors suggests that inclusion of the orbital degrees of freedom, which may lead to competing tendencies towards different magnetically ordered states, is more crucial than taking into account the strong correlations. Our findings further indicate that in order to properly model these materials, the electronic states away from the Fermi level have to be considered.
We review theoretical investigations on the origin of the orbital selective phase where localized and itinerant electrons coexist in the d shell at intermediate strength of the on-site Coulomb interactions between electrons. In particular, the effect of spatial fluctuations on the phase diagram of the two-orbital Hubbard model with unequal bandwidths is discussed. And different band dispersions in different orbitals as well as different magnetically ordered states in different orbitals which are responsible for orbital selective phase transitions are emphasized. This is due to the fact that these two mechanisms are independent of the Hund's rule coupling, and are completely distinct from other well-known mechanisms like orbitals of unequal bandwidths and orbitals with different degeneracies. Moreover, crystal field splitting is not required in these two recently proposed mechanisms.
Using the two-dimensional ionic Hubbard model as a simple basis for describing the electronic structure of silicene in the presence of an electric field induced by the substrate, we use the coherent-potential approximation to calculate the zero-temperature phase diagram and associated spectral function at half filling. We find that any degree of symmetry-breaking induced by the electric field causes the silicene structure to lose its Dirac fermion characteristics, thus providing a simple mechanism for the disappearance of the Dirac cone.
We propose that a combination of the semiclassical approximation with Monte Carlo simulations can be an efficient and reliable impurity solver for dynamical mean field theory equations and their cluster extensions with large cluster sizes. In order to show the reliability of the method, we consider two test cases: (i) the single-band Hubbard model within the dynamical cluster approximation with 4- and 8-site clusters and (ii) the anisotropic two-orbital Hubbard model with orbitals of different band width within the single-site dynamical mean field theory. We compare our results with those obtained from solving the dynamical mean field equations with continuous time and determinant quantum Monte Carlo. In both test cases we observe reasonable values of the metal-insulator critical interaction strength $U_c/t$ and, while some details of the spectral functions cannot be captured by the semiclassical approximation due to the freezing of dynamical fluctuations, the main features are reproduced by the approach.
We investigate the nature of the insulating phases in a bilayer Hubbard model with intralayer coupling $t$ and interlayer coupling $t_{\perp}$ at large interaction strength $U/t$ and half-filling. We consider a dynamical cluster approximation with a cluster size of $N_c=2\times4$, where short-range spatial fluctuations as well as on-site dynamical fluctuations are emphasized. By varying the band splitting ($t_{\perp}/t$), we find that at $t_{\perp}/t\simeq1.5$ the Mott behavior is rapidly suppressed in the momentum sectors ($π,0$) and ($0,π$). At $t_{\perp}/t\simeq2.5$ Mott features dominate in the momentum sectors ($π,π$) of the bonding band and ($0,0$) of the anti-bonding band and at $t_{\perp}/t\simeq3.0$ a tiny scattering rate is observed in all momentum sectors at the Fermi level, indicating a transition from a Mott to a band insulator. We attribute such a momentum-dependent evolution of the insulating behavior to the competition and cooperation between short-range spatial fluctuations and interlayer coupling $t_{\perp}$ with the help of the Coulomb interaction $U$. Finally, we also discuss the possible appearance of non-Fermi liquid behavior away from half-filling.
We study strong stability of Nash equilibria in load balancing games of m (m >= 2) identical servers, in which every job chooses one of the m servers and each job wishes to minimize its cost, given by the workload of the server it chooses. A Nash equilibrium (NE) is a strategy profile that is resilient to unilateral deviations. Finding an NE in such a game is simple. However, an NE assignment is not stable against coordinated deviations of several jobs, while a strong Nash equilibrium (SNE) is. We study how well an NE approximates an SNE. Given any job assignment in a load balancing game, the improvement ratio (IR) of a deviation of a job is defined as the ratio between the pre- and post-deviation costs. An NE is said to be a r-approximate SNE (r >= 1) if there is no coalition of jobs such that each job of the coalition will have an IR more than r from coordinated deviations of the coalition. While it is already known that NEs are the same as SNEs in the 2-server load balancing game, we prove that, in the m-server load balancing game for any given m >= 3, any NE is a (5/4)-approximate SNE, which together with the lower bound already established in the literature yields a tight approximation bound. This closes the final gap in the literature on the study of approximation of general NEs to SNEs in load balancing games. To establish our upper bound, we make a novel use of a graph-theoretic tool.
By calculating orbitally resolved Pauli susceptibilities within maximally localized Wannier orbital basis transformed from first principles band structures, we find that magnetism in Fe$_{1+x}$Te still has its itinerant origin even without Fermi surface nesting, provided orbital modulation of particle-hole excitations are considered. This leads to strong magnetic instabilities at wave vector (0,$π$)/($π$,0) in d$_{xz}$/d$_{yz}$ orbitals that are responsible for the bicollinear antiferromagnetic state as extra electrons donated from excess Fe are considered. Magnetic exchange coupling between excess Fe and in-plane Fe further stabilizes the bicollinear antiferromagnetic order. Our results reveal that magnetism and superconductivity in iron chalcogenides may have different orbital origin, as Pauli susceptibilities of different orbitals evolve differently as a function of concentration of excess Fe and height of the chalcogen atom measured from the iron plane.
Based on the analysis of a two-orbital Hubbard model within a mean-field approach, we propose a mechanism for an orbital selective phase transition (OSPT) where coexistence of localized and itinerant electrons can be realized. We show that this OSPT exists both at and near half filling even in the absence of crystal field splittings or when bandwidths, orbital degeneracies and magnetic states are equal for both orbitals provided the orbitals have different band dispersions. Such conditions should be generally satisfied in many materials. We find that this OSPT is not sensitive to the strength of Hund's rule coupling and that heavy doping favors the collinear antiferromagnetic state over the OSPT. We discuss our results in relation to the iron pnictides.
The anisotropic two-orbital Hubbard model with different bandwidths and degrees of frustration in each orbital is investigated in the framework of both single-site dynamical mean-field theory (DMFT) as well as its cluster extension (DCA) for clusters up to four sites combined with a continuous-time quantum Monte Carlo algorithm. This model shows a rich phase diagram which includes the appearance of orbital selective phase transitions, non-Fermi liquid behavior as well as antiferromagnetic metallic states. We discuss the advantages and drawbacks of employing the single-site DMFT with respect to DCA and the consequences for the physical picture obtained out of these calculations. Finally, we argue that such a minimal model may be of relevance to understand the nature of the antiferromagnetic metallic state in the iron-pnictide superconductors as well as the origin of the small staggered magnetization observed in these systems.
Motivated by the unexplored complexity of phases present in the multiorbital Hubbard model, we analyze in this work the behavior of a degenerate two-orbital anisotropic Hubbard model at half filling where both orbitals have equal bandwidths and one orbital is constrained to be paramagnetic (PM), while the second one is allowed to have an antiferromagnetic (AF) solution. Such a model may be relevant for a large class of correlated materials with competing magnetic states in different orbitals like the recently discovered Fe-based superconductors. Using a dynamical cluster approximation we observe that novel orbital selective phase transitions appear regardless of the strength of the Ising Hund's rule coupling $J_z$. Moreover, the PM orbital undergoes a transition from a Fermi liquid (FL) to a Mott insulator through a non-FL phase while the AF orbital shows a transition from a FL to an AF insulator through an AF metallic phase. We discuss the implications of the results in the context of the Fe-based superconductors.
We investigate the electronic and magnetic properties of the frustrated triangular-lattice antiferromagnets Cs$_2$CuCl$_4$ and Cs$_2$CuBr$_4$ in the framework of density functional theory. Analysis of the exchange couplings J and J' using the available X-ray structural data corroborates the values obtained from experimental results for Cs$_2$CuBr$_4$ but not for Cs$_2$CuCl$_4$. In order to understand this discrepancy, we perform a detailed study of the effect of structural optimization on the exchange couplings of Cs$_2$CuCl$_4$ employing different exchange-correlation functionals. We find that the exchange couplings depend on rather subtle details of the structural optimization and that only when the insulating state (mediated through spin polarization) is present in the structural optimization, we do have good agreement between the calculated and the experimentally determined exchange couplings. Finally, we discuss the effect of interlayer couplings as well as longer-ranged couplings in both systems.
By applying density functional theory, we find strong evidence for an itinerant nature of magnetism in two families of iron pnictides. Furthermore, by employing dynamical mean field theory with continuous time quantum Monte Carlo as an impurity solver, we observe that the antiferromagnetic metal with small magnetic moment naturally arises out of coupling between unfrustrated and frustrated bands. Our results point to a possible scenario for magnetism in iron pnictides where magnetism originates from a strong instability at the momentum vector ($π$, $π$, $π$) while it is reduced by quantum fluctuations due to the coupling between weakly and strongly frustrated bands.
The optical conductivity of LaFeAsO, BaFe$_2$As$_2$, SrFe$_2$As$_2$, and EuFe$_2$As$_2$ in the spin-density wave (SDW) state is investigated within density functional theory (DFT) in the framework of spin-polarized generalized gradient approximation (GGA) and GGA+U. We find a strong dependence of the optical features on the Fe magnetic moments. In order to recover the small Fe magnetic moments observed experimentally, GGA+$U_{\rm eff}$ with a suitable choice of negative on-site interaction $U_{\rm eff}=U-J$ was considered. Such an approach may be justified in terms of an overscreening which induces a relatively small U compared to the Hund's rule coupling J, as well as a strong Holstein-like electron-phonon interaction. Moreover, reminiscent of the fact that GGA+$U_{\rm eff}$ with a positive $U_{\rm eff}$ is a simple approximation for reproducing a gap with correct amplitude in correlated insulators, a negative $U_{\rm eff}$ can also be understood as a way to suppress magnetism and mimic the effects of quantum fluctuations ignored in DFT calculations. With these considerations, the resulting optical spectra reproduce the SDW gap and a number of experimentally observed features related to the antiferromagnetic order. We find electronic contributions to excitations that so far have been attributed to purely phononic modes. Also, an orbital resolved analysis of the optical conductivity reveals significant contributions from all Fe 3d orbitals. Finally, we observe that there is an important renormalization of kinetic energy in these SDW metals, implying that the effects of correlations cannot be neglected.
We investigate the phase diagram of a two-band frustrated Hubbard model in the framework of dynamical mean field theory. While a first-order phase transition occurs from a paramagnetic (PM) metal to an antiferromagnetic (AF) insulator when both bands are equally frustrated, an intermediate AF metallic phase appears in each band at different $U_c$ values if only one of the two bands is frustrated, resulting in continuous orbital-selective phase transitions from PM metal to AF metal and AF metal to AF insulator. We argue that such intermediate phases are possibly related to the puzzling AF metallic state with small magnetization observed in undoped iron-pnictide superconductors as well as to the pseudogap features observed in optical experiments.
We report a combined experimental and theoretical investigation of the layered antimonide PrMnSbO which is isostructural to the parent phase of the iron pnictide superconductors. We find linear resistivity near room temperature and Fermi liquid-like T^{2} behaviour below 150 K. Neutron powder diffraction shows that unfrustrated C-type Mn magnetic order develops below \sim 230 K, followed by a spin-flop coupled to induced Pr order. At T \sim 35 K, we find a tetragonal to orthorhombic (T-O) transition. First principles calculations show that the large magnetic moments observed in this metallic compound are of local origin. Our results are thus inconsistent with either the itinerant or frustrated models proposed for symmetry breaking in the iron pnictides. We show that PrMnSbO is instead a rare example of a metal where structural distortions are driven by f-electron degrees of freedom.
We investigate the effect of Na intercalation in the layered Mott insulator TiOCl within the framework of density functional theory. We show that the system remains always insulating for all studied Na concentrations, and the evolution of the spectral weight upon Na doping is consistent with recent photoemission experiments. We predict the Na-doped superlattice structures, and show that substitutions of O by F, Cl by S, or Ti by V (or Sc), respectively, fail to metallize the system. We propose a description in terms of a multiorbital ionic Hubbard model in a quasi-two-dimensional lattice and discuss the nature of the insulating state under doping. Finally, a likely route for metallizing TiOCl by doping is proposed.
Within the framework of density functional theory we investigate the nature of magnetism in various families of Fe-based superconductors. (i) We show that magnetization of stripe-type antiferromagnetic order always becomes stronger when As is substituted by Sb in LaOFeAs, BaFe$_2$As$_2$ and LiFeAs. By calculating Pauli susceptibilities, we attribute the magnetization increase obtained after replacing As by Sb to the enhancement of an instability at $(π,π)$. This points to a strong connection between Fermi surface nesting and magnetism, which supports the theory of the itinerant nature of magnetism in various families of Fe-based superconductors. (ii) We find that within the family LaOFe$Pn$ ($Pn$=P, As, Sb, Bi) the absence of an antiferromagnetic phase in LaOFeP and its presence in LaOFeAs can be attributed to the competition of instabilities in the Pauli susceptibility at $(π,π)$ and $(0,0)$, which further strengthens the close relation between Fermi surface nesting and experimentally observed magnetization. (iii) Finally, based on our relaxed structures and Pauli susceptibility results, we predict that LaOFeSb upon doping or application of pressure should be a candidate for a superconductor with the highest transition temperature among the hypothetical compounds LaOFeSb, LaOFeBi, ScOFeP and ScOFeAs while the parent compounds LaOFeSb and LaOFeBi should show at ambient pressure a stripe-type antiferromagnetic metallic state.