Phase diagrams of the one-dimensional extended Hubbard model (including nearest-neighbor interaction V ) at half-and quarter-filling are studied by observing level crossings of excitation spectra using the exact diagonalization. This method is based on the Tomonaga-Luttinger liquid theory including logarithmic corrections which stem from the renormalization of the Umklapp-and the backward-scattering effects. Using this approach, the phase boundaries are determined with high accuracy, and then the structure of the phase diagram is clarified. At half-filling, the phase diagram consists of two Berezinskii-Kosterlitz-Thouless (BKT) transition lines and one Gaussian transition line in the charge sector, and one spin-gap transition line. This structure reflects the U(1) ⊗ SU(2) symmetry of the electron system. Near the U = 2V line, the Gaussian and the spin-gap transitions take place independently from the weak-to the intermediate-coupling region, but these two transition lines are coupled in the strong-coupling region. This result demonstrates existence of a tricritical point and a bond-charge-density-wave (BCDW) phase between charge-and spin-densitywave (CDW, SDW) phases. To clarify this mechanism of the transition, we also investigate effect of a correlated hopping term which plays a role to enlarge BCDW and bond-spin-density-wave (BSDW) phases. At quarter-filling, a similar crossover phenomenon also takes place in the large-V region involving spin-gap and BKT-type metal-insulator transitions. One-dimensional (1D) electron systems have been extensively studied motivated not only by theoretical interest but also by the discovery of quasi-1D conductors and high-T c superconductivity. In the 1D electron systems, due to the charge-spin separation, the low-energy excitations in the charge and the spin sectors may have gaps independently, and then various phases can appear. However, phenomena caused by interplay between these two degrees of freedom have not been fully understood even in simple models. In this paper, we turn our attention to the phase transitions in the so-called extended Hubbard model (EHM),at half-and quarter-filling, where both charge and spin gaps can open. The EHM at half-filling has been studied using various approaches. In the weak-coupling limit, the phase diagram is analytically obtained by the g-ology 1,2,3,4 (see Appendix A). According to the result, there appear insulating charge-(CDW) and spin-density-wave (SDW) phases, and metallic phases where the singlet superconducting (SS) or the triplet superconducting (TS) correlation is dominant [see Fig. 1(b)]. On the other hand, in the strong-coupling limit, the perturbation theory gives the phase boundary of the CDW-SDW transition 5,6,7 and of the phase separation. 8,9,10 The rest of region has been discussed by numerical analysis. 6,11,12,13,14,15 However, the phase diagrams are not fully understood, because the charge and the spin gaps open exponentially slow [see Eqs. (18) and (21)], which makes it difficult to determine the phase boundaries b...

The electronic structure of the La2−xSrxCuO4 (LSCO) system has been studied by angle-resolved photoemission spectroscopy (ARPES). We report on the evolution of the Fermi surface, the superconducting gap and the band dispersion around the extended saddle point k = (π, 0) with hole doping in the superconducting and metallic phases. As hole concentration x decreases, the flat band at (π, 0) moves from above the Fermi level (EF) for x > 0.2 to below EF for x < 0.2, and is further lowered down to x = 0.05. From the leading-edge shift of ARPES spectra, the magnitude of the superconducting gap around (π, 0) is found to monotonically increase as x decreases from x = 0.30 down to x = 0.05 even though Tc decreases in the underdoped region, and the superconducting gap appears to smoothly evolve into the normal-state gap at x = 0.05. It is shown that the energy scales characterizing these low-energy structures have similar doping dependences. For the heavily overdoped sample (x = 0.30), the band dispersion and the ARPES spectral lineshape are analyzed using a simple phenomenological self-energy form, and the electronic effective mass enhancement factor m * /m b ≃ 2 has been found. As the hole concentration decreases, an incoherent component that cannot be described within the simple self-energy analysis grows intense in the high-energy tail of the ARPES peak. Some unusual features of the electronic structure observed for the underdoped region (x < ∼ 0.10) are consistent with the numerical works on the stripe model.

The phase transition between charge- and spin-density-wave (CDW, SDW) phases is studied in the one-dimensional extended Hubbard model at half-filling. We discuss whether the transition can be described by the Gaussian and the spin-gap transitions under charge-spin separation, or by a direct CDW-SDW transition. We determine these phase boundaries by level crossings of excitation spectra which are identified according to discrete symmetries of wave functions. We conclude that the Gaussian and the spin-gap transitions take place separately from weak- to intermediate-coupling region. This means that the third phase exists between the CDW and the SDW states. Our results are also consistent with those of the strong-coupling perturbative expansion and of the direct evaluation of order parameters.Comment: 5 pages(REVTeX), 5 figures(EPS), 1 table, also available from http://wwwsoc.nacsis.ac.jp/jps/jpsj/1999/p68a/p68a42/p68a42h/p68a42h.htm

We report on the result of angle-resolved photoemission (ARPES) study of La2−xSrxCuO4 (LSCO) from an optimally doped superconductor (x = 0.15) to an antiferromagnetic insulator (x = 0). Near the superconductor-insulator transition (SIT) x ∼ 0.05, spectral weight is transferred with hole doping between two coexisting components, suggesting a microscopic inhomogeneity of the dopedhole distribution. For the underdoped LSCO (x ≤ 0.12), the dispersive band crossing the Fermi level becomes invisible in the (0, 0)−(π, π) direction unlike Bi2Sr2CaCu2O8+y. These observations may be reconciled with the evolution of holes in the insulator into fluctuating stripes in the superconductor.PACS numbers: 74.25.Jb, 71.30.+h, 74.72.Dn, 74.62.Dh The key issue to clarify the nature of high-temperature superconductivity in the cuprates is how the electronic structure evolves from the antiferromagnetic insulator (AFI) to the superconductor (SC) with hole doping. For the hole-doped CuO 2 planes in the superconductors, band dispersions and Fermi surfaces have been extensively studied by angle-resolved photoemission spectroscopy (ARPES) primarily on Bi 2 Sr 2 CaCu 2 O 8+y (BSCCO) 1-4 . Also for the undoped AFI, band dispersions have been observed for Sr 2 CuO 2 Cl 2 5,6 . However, the band structures of the AFI and the SC are distinctly different and ARPES data have been lacking around the boundary between the AFI and the SC. In order to reveal the missing link, the present ARPES study has been performed on La 2−x Sr x CuO 4 (LSCO), which covers continuously from the SC to the AFI in a single system.In addition, the family of LSCO systems show a suppression of T c at a hole concentration δ ≃ 1/8, while the BSCCO system does not. As the origin of the anomaly at δ ≃ 1/8, the instability towards the spin-charge order in a stripe form has been extensively discussed on the basis of the incommensurate peaks in inelastic neutron scattering (INS) 7-9 . Comparing the ARPES spectra of LSCO and BSCCO will help us to clarify the impact of the stripe fluctuations.In the present paper, we discuss the novel observation of two spectral components coexisting around the SIT (x ∼ 0.05), the unusual disappearance of the Fermi surface near (π/2, π/2) in the underdoped LSCO (x ≤ 0.12) 4 , and their relevance to the stripe fluctuations.The ARPES measurements were carried out at beamline 5-3 of Stanford Synchrotron Radiation Laboratory (SSRL). Incident photons had an energy of 29 eV and were linearly polarized. The total energy resolution was approximately 45 meV and the angular resolution was ±1 degree. Single crystals of LSCO were grown by the traveling-solvent floating-zone method and then annealed so that the oxygen content became stoichiometric. The accuracy of the hole concentration δ was ±0.01. The x = 0 samples were slightly hole doped by excess oxygen so that δ ≃ 0.005 was deduced from its Néel temperature T N = 220 K 10 . The spectrometer was kept in an ultra high vacuum better than 5 × 10 −11 Torr during the measurements. The samples were cleaved ...

High resolution angle-resolved photoemission measurements have been carried out on (La(1.4--x)-Nd(0.6)Sr(x))CuO(4), a model system with static one-dimensional (1D) charge ordering (stripe), and (La(1.85)-Sr(0.15))CuO(4), a high temperature superconductor (T(c) = 40 K) with possible dynamic stripes. In addition to the straight segments near ( pi,0) and ( 0,pi) antinodal regions, we have identified the existence of spectral weight along the [1,1] nodal direction in the electronic structure of both systems. This observation of nodal state, together with the straight segments near antinodal regions, reveals the dual nature of the electronic structure of stripes due to the competition of order and disorder.