The t–J model with a prohibition of “doubles” (doubly occupied sites) in the lower Hubbard band and holes in the upper band is considered in the regime of strong electron correlations. The specific features of the electronic states are described. An expression is obtained for the transverse dynamic susceptibility χ as a function of wave vector q and frequency ω in the random phase approximation (RPA) in the metallic state of the system, and the static susceptibility (ω=0) is analyzed in the limits q=0 and q=Q=(π,π) in a square lattice as a function of the electron density n. The insulating limit (n→1) is described as a spin liquid (SL) state. An expression for the magnetic susceptibility is proposed which unifies the RPA and the SL concept in the presence of site percolation. It is shown that in a certain concentration region the static susceptibility is negative (i.e., a diamagnetic effect), which indicates the possibility of a superconducting state with an exchange mechanism of pairing.
The t–J model with a prohibition of “doubles” (doubly occupied sites) in the lower Hubbard band and holes in the upper band is considered in the regime of strong electron correlations. The specific features of the electronic states are described. An expression is obtained for the transverse dynamic susceptibility χ as a function of wave vector q and frequency ω in the random phase approximation (RPA) in the metallic state of the system, and the static susceptibility (ω=0) is analyzed in the limits q=0 and q=Q=(π,π) in a square lattice as a function of the electron density n. The insulating limit (n→1) is described as a spin liquid (SL) state. An expression for the magnetic susceptibility is proposed which unifies the RPA and the SL concept in the presence of site percolation. It is shown that in a certain concentration region the static susceptibility is negative (i.e., a diamagnetic effect), which indicates the possibility of a superconducting state with an exchange mechanism of pairing.
The Hubbard model in the limit of strong Coulomb interaction (the (t-J) model) is studied by the diagram technique for Hubbard operators. The generalized random-phase approximation (GRPA) is formulated as an approximation taking into account electron loop-type diagrams. Within the framework of this approximation, both the dynamic magnetic and dielectric susceptibilities are calculated for a wide interval of electron concentrations, 0<n<1. The magnetic susceptibility shows a crossover in the magnetic behaviour of the system from pure itinerant magnetism to magnetism with localized magnetic moments. The crossover occurs at a critical point, nc, when the chemical potential for electrons of the lower Hubbard band changes sign. Magnetic phase diagrams are constructed on the (t/U, n) plane at T=0 for different types of crystal lattices. The behaviour of magnetic phases with temperature is also studied. The GRPA leads to a too drastic crossover, especially at low temperatures, because of insufficient allowance for the charge and spin fluctuations on a site. Summation of special diagram series shows that the Gaussian fluctuations of the effective electric and magnetic fields lead to a smoother picture of the crossover.
The properties of spin excitations superposed on a uniform ground state with antiferromagnetic (or spiral) spin structure are studied in a 2 D Hubbard model. Expressions are derived for the spin susceptibility in the random phase approximation (RPA) using split Hubbard bands as a zeroth approximation. The calculated collective modes with dispersion ω ( Q )= c | Q −(π, π)| near Q ∼(π, π) reproduce well the characteristics of the spin excitations observed in undoped cuprates. For doped systems with an antiferromagnetic structure of the ground state, calculating X ″( Q,ω →0) gives the same mode with a peak at Q ∼(π, π), regardless of the type of Fermi surface. It is shown that in doped systems with a spiral ground state spin structure, X ″( Q,ω →0) peaks occur with incommensurate quasimomenta Q that are coupled to the spirality vector.
The dynamical spin susceptibility χ( q , ω), of the t-J model in two-dimensional square lattice have been studied based on the mean field approximation to the slave-boson formalism. The static spin susceptibility, χ( q , 0), and Im χ( q , ω) at various energy, ω, have been calculated. We found; 1) χ(0, 0) has only weak temperature dependence 2) T 1 T ∝( T +θ) for T≳0.1J with θ linearly dependent on the doping rate, δ 3) T 1 T has a minimum at T =(0.1∼0.2) J 4) Im χ( q , ω) at q = Q ≡(π, π) has a gap like structure as a function of ω if the temperature is low. All these are consistent with experimental results.
The dynamical spin susceptibility χ(q, ω), of the t-J model in two-dimensional square lattice have been studied based on the mean field approximation to the slave-boson formalism. The static spin susceptibility, χ(q, 0), and Im χ(q, ω) at various energy, ω, have been calculated. We found; 1) χ(0, 0) has only weak temperature dependence 2) T1T ∝ (T+θ) for T≳0.1J with θ linearly dependent on the doping rate, δ 3) T1T has a minimum at T= (0.1∼0.2)J 4) Im χ(q, ω) at q = Q = (π, π) has a gap like structure as a function of ω if the temperature is low. All these are consistent with experimental results.
The dynamical spin susceptibility χ( q , ω), of the t-J model in two-dimensional square lattice have been studied based on the mean field approximation to the slave-boson formalism. The static spin susceptibility, χ( q , 0), and Im χ( q , ω) at various energy, ω, have been calculated. We found; 1) χ(0, 0) has only weak temperature dependence 2) T 1 T ∝( T +θ) for T≳0.1J with θ linearly dependent on the doping rate, δ 3) T 1 T has a minimum at T =(0.1∼0.2) J 4) Im χ( q , ω) at q = Q ≡(π, π) has a gap like structure as a function of ω if the temperature is low. All these are consistent with experimental results.
We consider the slave-fermion--Schwinger-boson decomposition of a t-J model for a two-dimensional antiferromagnet in which the next-nearest-neighbor hopping is taken into account by an explicit t' term. This technique is particularly suitable to describe systems with low hole density, as in the parent phases of high-Tc superconducting (HTSC) materials, and we examine the experimental consequences of the model. Using mean-field states which admit finite spiral and off-diagonal order, we compute first the dynamic susceptibility \ensuremath{\chi}(q,\ensuremath{\omega}) from fluctuations in the boson (spin) degrees of freedom, and investigate a selected parameter regime for the model to seek features consistent with inelastic neutron scattering experiments on underdoped HTSC crystals. We find peaks in the scattered intensity at incommensurate wave vectors (\ensuremath{\pi}\ifmmode\pm\else\textpm\fi{}2k0,\ifmmode\pm\else\textpm\fi{}\ensuremath{\pi}) and (\ensuremath{\pi},\ifmmode\pm\else\textpm\fi{}\ensuremath{\pi}\ifmmode\pm\else\textpm\fi{}2k0), but within our approximation the q-integrated susceptibility as a function of frequency and temperature resembles that from spin-wave theory, which does not contain the scaling behavior over all values of \ensuremath{\omega}/T found experimentally. We show also the form of the nuclear magnetic resonance relaxation time ${\mathit{T}}_{1}^{\mathrm{\ensuremath{-}}1}$(T), which measures the zero-frequency limit of the susceptibility. The imaginary part of the physical electron Green function is a convolution of both fermionic (hole) and bosonic Green functions, and so incorporates the effect of spin dynamics on the charge motion. We calculate the photoemitted intensity, a direct experimental measurement of this quantity, focusing on its behavior near the hole pockets at (\ifmmode\pm\else\textpm\fi{}\ensuremath{\pi}/2,\ifmmode\pm\else\textpm\fi{}\ensuremath{\pi}/2) which characterize this type of model. The peak photocurrent occurs not at the hole Fermi surface, but at points separated from it by k0 due to the boson incommensuration.