Ab initiocalculation of impurity effects in copper oxide materials

Abstract

We describe a method for calculating, within density functional theory, the electronic structure associated with typical defects which substitute for Cu in the CuO 2 planes of high-T c superconducting materials. The focus is primarily on Bi 2 Sr 2 CaCu 2 O 8 , the material on which most STM measurements of impurity resonances in the superconducting state have been performed. The magnitudes of the effective potentials found for Zn, Ni and vacancies on the in-plane Cu sites in this host material are remarkably consistent with phenomenological fits of potential scattering models to STM resonance energies. The effective potential ranges are quite short, of order 1 Å with weak long range tails, in contrast to some current models of extended potentials which attempt to fit STM data.For the case of Zn and Cu vacancies, the effective potentials are strongly repulsive, and states on the impurity site near the Fermi level are simply removed. The local density of states (LDOS) just above the impurity is nevertheless found to be a maximum in the case of Zn and a local minimum in case of the vacancy, in agreement with experiment. The Zn and Cu vacancy patterns are explained as due to the long-range tails of the effective impurity potential at the sample surface. The case of Ni is richer due to the Ni atom's strong hybridization with states near the Fermi level; in particular, the short range part of the potential is attractive, and the LDOS is found to vary rapidly with distance from the surface and from the impurity site. We propose that the current controversy surrounding the observed STM patterns can be resolved by properly accounting for the effective impurity potentials and wave-functions near the cuprate surface. Other aspects of the impurity states for all three species are discussed.2

Interpretation of Scanning Tunneling Quasiparticle Interference and Impurity States in Cuprates

“…So far, these theories have been expressed entirely in terms of phenomenological effective hoppings in the Cu tightbinding model. First principles calculations for Zn in BSCCO in the normal state [27] provide some evidence in support of the filter picture, but until recently it was not possible to include both superconductivity and the various atomic wave functions extending into the barrier layers responsible for the filter. Nieminen et al investigated the conductance spectrum in the BSCCO system using an analysis based on atomiclike wave functions [28], and showed that for the homogeneous system it could be decomposed in a series of tunneling paths, as postulated by the earlier crude proposals [25,26].…”

“…1(c).As pointed out in Refs. [25][26][27][28], the problem lies in the consideration of the Cu lattice sites far from the BiO surface. The correct quantity to study is the continuum LDOS ρ(r, Ω 0 ) at the height of the STM tip, which we assume to be at z = 5Å above the BiO surface.…”

Insensitivity ofd-wave pairing to disorder in the high-temperature cuprate superconductors

“…Because the screened Coulomb potential due to these defects is very short range [11], such impurities are frequently modeled by pointlike (δ-function) scatterers. The expected form for the suppression of superconductivity in the BCS theory of d-wave superconductors is then identical to the expression given by Abrikosov and Gor'kov [12] for magnetic impurities in s-wave superconductors (see, e.g., Ref.13).…”

Local Electronic Structure ofBi2Sr2CaCu2O8near Oxygen Dopants: A Window on the High-THe1, Nunner2, Hirschfeld3 et al. 2006Phys. Rev. Lett.Self Cite549510View full textAdd to dashboardCite

“…On the other hand, the bands and Fermi surface given by LDA agree relatively well with ARPES results on optimally doped BSCCO-2212, with minor exceptions (see below). Furthermore, as argued in [23], we are primarily interested in high-energy, localized impurity states which should be little affected by the strong electronic correlations which renormalize the band states near the Fermi surface.…”

“…We use the exchange-correlation functional determined by Ceperly and Alder [21] and parameterized by Perdew and Zunger [22]. Details of the application of the method to impurity problems in Bi 2 Sr 2 CaCu 2 O 8 have been given in [23]. It is known that LDA fails to give the correct band structure for cuprates in general, and in particular fails to describe the metallic/insulating character of the parent compounds.…”