arXiv subjects
Publications and source records attributed to Adolfo Avella.
The Composite Operator Method (COM) is formulated, its internals illustrated in detail and some of its most successful applications reported. COM endorses the emergence, in strongly correlated systems (SCS), of composite operators, optimally deals with their unusual features and implements algebra constraints, and other relevant symmetries, in order to properly compute the unconventional properties of SCS.
We analyze the two-orbital Hubbard model by means of the Composite Operator Method with the aim at studying the phenomenon of orbital selective Mott transition (OSMT). The model contains an interorbital interaction $U'$, in addition to the usual intraorbital one $U$. As warming-up approximation, we use a basis of two operators only, the Hubbard operators. The analysis of the density of states at the chemical potential as a function of the ratio between the bandwidths of the two orbitals shows the clear signature of an orbital selective Mott transition as expected.
By means of the Density Matrix Renormalization Group technique, we have studied the region where $XXZ$-like behavior is most likely to emerge within the phase diagram of the F-AF anisotropic extended ($J-J'$) Heisenberg chain. We have analyzed, in great detail, the equal-time two-spin correlation functions, both in- and out-of- plane, as functions of the distance (and momentum). Then, we have extracted, through an accurate fitting procedure, the exponents of the asymptotic power-law decay of the spatial correlations. We have used the exact solution of $XXZ$ model ($J'=0$) to benchmark our results, which clearly show the expected agreement. A critical value of $J'$ has been found where the relevant power-law decay exponent is independent of the in-plane nearest-neighbor coupling.
We present a study of the entanglement properties of the F-AF zig-zag Heisenberg chain done by means of the Density Matrix Renormalization Group method. In particular, we have selected the concurrence as measure of entanglement and checked its capability to signal the presence of quantum phase transitions within the previously found ergodicity phase diagram [E. Plekhanov, A. Avella, and F. Mancini, Phys. Rev. B \textbf{74}, 115120 (2006)]. By analyzing the behavior of the concurrence, we have been able not only to determine the position of the transition lines within the phase diagram of the system, but also to identify a well defined region in the parameter space of the model that shows a complex spin ordering indicating the presence of a new phase of the system.
We present a concise, but systematic, review of the ergodicity issue in strongly correlated systems. After giving a brief historical overview, we analyze the issue within the Green's function formalism by means of the equations of motion approach. By means of this analysis, we are able to individuate the primary source of non-ergodic dynamics for a generic operator and also to give a recipe to compute unknown quantities characterizing such a behavior within the Composite Operator Method. Finally, we present examples of non-trivial strongly correlated systems where it is possible to find a non-ergodic behavior.
We present the Composite Operator Method (COM) as a modern approach to the study of strongly correlated electronic systems, based on the equation of motion and Green's function method. COM uses propagators of composite operators as building blocks at the basis of approximate calculations and algebra constrains to fix the representation of Green's functions in order to maintain the algebraic and symmetry properties.
We study the spin-$\frac32$ Hubbard-Kondo lattice model by means of the Composite Operator Method, after applying a Holstein-Primakov transformation. The spin and particle dynamics in the ferromagnetic state are calculated by taking into account strong on-site correlations between electrons and antiferromagnetic exchange among $\frac32$ spins, together with usual Hund coupling between electrons and spins.
The charge and spin sectors, which are intimately coupled to the fermionic one, of the $t$-$t'$ Hubbard model have been computed self-consistently within the two-pole approximation. The relevant unknown correlators appearing in the causal bosonic propagators have been computed by enforcing the constraints dictated by the hydrodynamics and the algebra of the composite operators coming into play. The proposed scheme of approximation extends previous calculations made for the fermionic sector of the $t$-$t'$ Hubbard model and the bosonic sector of the Hubbard model, which showed to be very effective to describe the overdoped region of cuprates (the former) and the magnetic response of their parent compounds (the latter).
A fully self-consistent calculation of the bosonic dynamics of the Hubbard model is developed within the Composite Operator Method. From one side we consider a basic set of fermionic composite operators (Hubbard fields) and calculate the retarded propagators. On the other side we consider a basic set of bosonic composite operators (charge, spin and pair) and calculate the causal propagators. The equations for the Green's functions (GF) (retarded and causal), studied in the polar approximation, are coupled and depend on a set of parameters not determined by the dynamics. First, the pair sector is self-consistently solved together with the fermionic one and the zero-frequency constants (ZFC) are calculated not assuming the ergodic value, but fixing the representation of the GF in such a way to maintain the constrains required by the algebra of the composite fields. Then, the scheme to compute the charge and spin sectors, ZFCs included, is given in terms of the fermionic and pair correlators.
We present the results of exact diagonalization calculations of the isolated and isothermal on-site static susceptibilities in the anisotropic extended Heisenberg model on a linear chain with periodic boundary conditions. Based on the ergodicity considerations we conclude that the isothermal susceptibility will diverge as $T\to 0$ both in finite clusters and in the bulk system in two non-ergodic regions of the phase diagram of the system.
The two-dimensional Hubbard model is studied within the Composite Operator Method (COM) with the residual self-energy computed in the Self-Consistent Born Approximation (SCBA). COM describes interacting electrons in terms of the new elementary excitations appearing in the system owing to strong correlations; residual interactions among these excitations are treated within the SCBA. The anomalous features appearing in the spectral function A(k,ω), the momentum distribution function n(k) and the Fermi surface are analyzed for various values of the filling (from overdoped to underdoped region) in the intermediate coupling regime at low temperatures. For low doping, in contrast with the ordinary Fermi-liquid behavior of a weakly-correlated metal found at high doping, we report the opening of a pseudogap and some non-Fermi-liquid features as measured for cuprates superconductors. In addition, we show the presence of kinks in the calculated electronic dispersion in agreement with ARPES data.
The issue of ergodicity is often underestimated. The presence of zero-frequency excitations in bosonic Green's functions determine the appearance of zero-frequency momentum-dependent quantities in correlation functions. The implicit dependence of matrix elements make such quantities also relevant in the computation of susceptibilities. Consequently, the correct determination of these quantities is of great relevance and the well-established practice of fixing them by assuming the ergodicity of the dynamics is quite questionable as it is not justifiable a priori by no means. In this manuscript, we have investigated the ergodicity of the dynamics of the $z$-component of the spin in the 1D Heisenberg model with anisotropic nearest-neighbor and isotropic next-nearest-neighbor interactions. We have obtained the zero-temperature phase diagram in the thermodynamic limit by extrapolating Exact and Lanczos diagonalization results computed on chains with sizes $L = 6 ÷26$. Two distinct non-ergodic regions have been found: one for $J^\prime/J_z \lesssim 0.3$ and $|J_\perp|/J_z < 1$ and another for $J^\prime/J_z \lesssim 0.25$ and $|J_\perp|/J_z = 1$. On the contrary, finite-size scaling of $T \neq 0$ results, obtained by means of Exact diagonalization on chains with sizes $L = 4 ÷18$, indicates an ergodic behavior of dynamics in the whole range of parameters.
The charge and spin dynamics of the two-dimensional Hubbard model in the paramagnetic phase is first studied by means of the two-pole approximation within the framework of the Composite Operator Method. The fully self-consistent scheme requires: no decoupling, the fulfillment of both Pauli principle and hydrodynamics constraints, the simultaneous solution of fermionic and bosonic sectors and a very rich momentum dependence of the response functions. The temperature and momentum dependencies, as well as the dependency on the Coulomb repulsion strength and the filling, of the calculated charge and spin susceptibilities and correlation functions are in very good agreement with the numerical calculations present in the literature.
The recently developed energy-scale-dependent Composite Operator Method is applied to the single-impurity Anderson model. A fully self-consistent solution is given and analyzed. At very low temperatures, the density of states presents, on the top of the high-energy background, a Kondo-like peak whose parameter dependence is discussed in detail. The proposed method reproduces the exact results known in the literature with very low numerical effort and it is applicable for arbitrary values of the external parameters.
The half-filled extended Hubbard model, in one and two dimensions, is studied by means of the 2-pole approximation within the Composite Operator Method with the aim at improving the possibilities to describe some of the experimental features observed for quasi-1D organic superconductors and Cu-O planes of cuprates. The phase diagrams ($T$-$V$ and $V$-$U$) are analyzed with respect to the paramagnetic metal - paramagnetic insulator - charge ordered phase transitions. The relevant features of the diagrams (rank of the phase transitions, critical points, reentrant behavior) are discussed in detail.
The electronic states of the two-dimensional Hubbard model are investigated by means of a 4-pole approximation within the Composite Operator Method. In addition to the conventional Hubbard operators, we consider other two operators which come from the hierarchy of the equations of motion and carry information regarding nearest-neighbor spin and charge configurations. By means of this operatorial basis, we can study the physics related to the energy scale of J=4t^2/U in addition to the one of U. Present results show relevant physical features, well beyond those previously obtained by means of a 2-pole approximation, such as a four-band structure with shadow bands and a quasi-particle peak at the Fermi level. The Fermi level stays pinned to the band flatness located at (pi,0)-point within a wide range of hole-doping (0 <= delta <= 0.15). A comprehensive analysis of double occupancy, internal energy, specific heat and entropy features have been also performed. All reported results are in excellent agreement with the data of numerical simulations.
The electronic states of the two-dimensional Hubbard model are investigated by means of a 4-pole approximation within the Composite Operator Method. In addition to the conventional Hubbard operators, we consider other two operators, which come from the hierarchy of the equations of motion. These operators carry information regarding surrounding spin and charge configurations. By means of this operatorial basis, we can study the physics related to the energy scale J=4t^2/U. Results present, in addition to the main two bands, a quasi-particle peak at the Fermi level, shadow bands and band flatness at (pi, 0) point.
We study the one- and two- dimensional extended Hubbard model by means of the Composite Operator Method within the 2-pole approximation. The fermionic propagator is computed fully self-consistently as a function of temperature, filling and Coulomb interactions. The behaviors of the chemical potential (global indicator) and of the double occupancy and nearest-neighbor density- density correlator (local indicators) are analyzed in detail as primary sources of information regarding the instability of the paramagnetic (metal and insulator) phase towards charge ordering driven by the intersite Coulomb interaction. Very rich phase diagrams (multiple first and second order phase transitions, critical points, reentrant behavior) have been found and discussed with respect to both metal-insulator and charge ordering transitions: the connections with the experimental findings relative to some manganese compounds are analyzed. Moreover, the possibility of improving the capability of describing cuprates with respect to the simple Hubbard model is discussed through the analysis of the Fermi surface and density of states features. We also report about the specific heat behavior in presence of the intersite interaction and the appearance of crossing points.