Dispersion of ordered stripe phases in the cuprates

Abstract

A phase separation model is presented for the stripe phase of the cuprates, which allows the doping dependence of the photoemission spectra to be calculated. The idealized limit of a well-ordered array of magnetic and charged stripes is analyzed, including effects of long-range Coulomb repulsion. Remarkably, down to the limit of two-cell wide stripes, the dispersion can be interpreted as essentially a superposition of the two end-phase dispersions, with superposed minigaps associated with the lattice periodicity. The largest minigap falls near the Fermi level; it can be enhanced by proximity to a (bulk) Van Hove singularity. The calculated spectra are dominated by two features -this charge stripe minigap plus the magnetic stripe Hubbard gap. There is a strong correlation between these two features and the experimental photoemission results of a two-peak dispersion in La2−xSrxCuO4, and the peak-dip-hump spectra in Bi2Sr2CaCu2O 8+δ . The differences are suggestive of the role of increasing stripe fluctuations. The 1/8 anomaly is associated with a quantum critical point, here expressed as a percolation-like crossover. A model is proposed for the limiting minority magnetic phase as an isolated two-leg ladder.

Phase separation models for cuprate stripe arrays

“…Therefore, the fact that the pseudogap closes in the overdoped regime in the cuprates strongly suggests that the stripe phase terminates at the same doping. Direct evidence for this has recently appeared 9,6,80,3 . Figure 14a compares the intensity I of the inelastic neutron peaks near (π, π) as a function of doping in YBCO 81 and LSCO 2,1 .…”

“…Thus, Uchida, et al 24 , studying the Hall coefficient R H , find a crossover from one-dimensional behavior (R H → 0 as T → 0) for x < x cr = 1/8 to two dimensional behavior (coupled charged stripes) for x > x cr . In YBCO, the spin gap grows slowly with doping for x < x cr , then more rapidly for x > x cr ; this behavior can be understood in terms of coupled spin ladders, as the coupling changes with the width of the charged stripes 9 . In Eu substituted LSCO 25 , the Meissner fraction is negligibly small for x < x cr , then grows roughly linearly with doping until x ≃ 0.18, staying large up to at least x = 0.22.…”

“…The dot-dashed curve shows that in the phase separation regime, the low energy physics can be approximated by the form of free energy assumed in Ref. 9,…”

“…For this purpose we replot the data as integrated spectral weight over a finite energy cut within energy ∆E of the Fermi surface. Figure 11a shows a cut with ∆E = 200meV, for the model of a 1/8 doped stripe array (i.e., x = 0.125) 9 . The pattern is readily understood: the stripe superlattice leads to a number of quasi-one-dimensional bands; however, due to structure factor effects they have significant intensity only near the original Fermi surface, solid line in Fig.…”

“…[7][8][9][10][11] While fluctuations 7,8 play an important role in the real cuprates, we have constructed ordered stripe arrays, for which detailed tight-binding calculations are possible 9 . Such calculations can aid in elucidating the structure of the charged stripes, both for wide stripes -what stabilizes the preferred hole density -and for narrow stripes -how do quantum size effects (QSE) modify properties of the stripes.…”