Van Hove exciton-cageons and high-Tc superconductivity XB. Polaronic coupling in the doped material

Abstract

A purely ionic interpretation of the tilting mode instabilities in La2−xAxCuO4 (A=Sr,Ba) is shown to be not selfconsistent: the dominant factor influencing the doping dependence of the interlayer mismatch is the large change in the Cu-O bond length. But such a large dependence of bond length on valence leads to a strong electron-phonon coupling, contradicting the assumption that covalent effects were negligible. This coupling is closely related to the vHs-JT effect. This new insight clarifies the role of the tilt-mode instabilities. The main JT coupling is not to these modes, but to the in-plane O-O bond stretching modes which split the vHs degeneracy. However, as these modes soften, they couple to the lower-lying tilt modes, so that the ultimate instability has a finite tilt component.This in turn simplifies the description of the JT instabilities. The tilt modes, while strongly coupled to the electrons, are quadratic in the tilt angle. On the other hand, the bond stretch modes have a large, linear coupling to electrons, with clear polaronic effects. This is connected with the fact that, near a vHs, valence fluctuations are slow and relatively long lived, coupling to local phonon modes. A striking result of this is that there will be a large polaronic band narrowing near the vHs, whether or not the vHs is near the Fermi level. This vHs-localized band narrowing provides a natural explanation for the common occurence of extended vHs's.

A survey of the Van Hove scenario for high-tc superconductivity with special emphasis on pseudogaps and striped phases

“…) The microscopic origin of the extended VHS's remains unclear, although there are a number of theoretical possibilities. One can generate extremely flat bands by renormalizing the band dispersion near a VHS by, e.g., strong electron-phonon coupling [150], as discussed in Section VIII.D (see Fig. 37) or in the presence of a (magnetic) pseudogap [111].…”

Flux phase as a dynamic Jahn-Teller phase: Berryonic matter in the cuprates

“…(The detailed interpretation involves partial hole localization on oxygens, and the large size difference between O − and O 2− . 35 ) Similar effects arise in other cuprates. Thus, high energy phonons behave similarly in LSCO and YBCO, but at low frequencies there are very different structural distortion.…”

“…While there are alternative interpretations 33 , it has been proposed 18 that this is the conventional renormalization associated with strong electron-phonon coupling 34 to the breathing mode phonons near this energy. Such renormalization can produce the extended saddle points observed near (π, 0) 35 .…”

“…Note the anticrossing behavior evident on several branches.) The pattern of distortion for a number of modes of interest is also shown, including the breathing 27 , half-breathing 20,81 (at (π, 0)), dimer 83 , quadrupolar 13 , ferroelectric 81 , and shear 35 modes. This by no means exhausts all relevant modes.…”

“…Such VHS splitting has recently received renewed interest 45,46 . We have suggested that the LTT tilt is a secondary modification of a shear mode instability due to lattice mismatch effects 35 : the underlying instability is to a shear (making inequivalent bondlengths), but the CuO 2 planes are under com-pression, so the LTT tilt arises as a compromise, making the Cu-O bondlengths inequivalent, without increasing the net Cu-Cu bondlength. Qualitatively, the LTT tilt pattern arises because the CuO 6 octahedra tilt as rigid units 47 : since the octahedra are corner sharing, alternate octahedra must tilt in opposite directions, leading to a distortion at wave vector Q = (π, π).…”