Quasiparticles and anomalous temperature dependence of the low-lying states in the colossal magnetoresistant oxide La2-2xSr1+2xMn2O7 (x=0.36) from angle-resolved photoemission

After years of research into colossal magnetoresistant (CMR) manganites using bulk techniques, there has been a recent upsurge in experiments directly probing the electronic states at or near the surface of the bilayer CMR materials La2-2xSr1+2xMn2O7 using angle-resolved photoemission or scanning probe microscopy. Here, we report temperature-dependent, angle-resolved photoemission data from single crystals with a doping level of x=0.36. The first important result is that there is no sign of a pseudogap in the charge channel of this material for temperatures below the Curie temperature T-C. The data show unprecedented sharp spectral features, enabling the unambiguous identification of clear, resolution-limited quasiparticle features from the bilayer split 3d(x)(2)-(2)(y)-derived Fermi surfaces both at the zone-face and zone diagonal k(F) locations. The data show that these low temperature Fermi surfaces describe closed shapes in k(parallel to), centered at the (pi/a,pi/a) points in the two dimensional Brillouin zone, and are not open and arclike in nature. The second important result concerns the temperature dependence of the electronic states. The spectra display strong incoherent intensity at high binding energies and a very strong temperature dependence, both characteristics reminiscent of polaronic systems. However, the clear and strong quasiparticle peaks at low temperatures are difficult to place within a polaronic scenario. A careful analysis of the temperature-dependent changes in the Fermi surface spectra both at the zone face and zone diagonal regions in k space indicates that the coherent quasiparticle weight disappears for temperatures significantly above T-C and that the k dependence of the T-induced changes in the spectra invalidates an interpretation of these data in terms of the superposition of a universal metallic spectrum and an insulating spectrum whose relative weight changes with temperature. In this sense, our data are not compatible with a phase separation scenario.