We outline a procedure for obtaining the electron-phonon spectral density by inversion of optical conductivity data, a process very similar in spirit to the McMillan-Rowell inversion of tunelling data. We assume both electron-impurity (elastic) and electron-phonon (inelastic) scattering processes. This procedure has the advantage that it can be utilized in the normal state. Furthermore, a very good qualitative result can be obtained explicitly, without iteration. We illustrate this technique on recently acquired far-infrared data in K3C60. We show that the electron-phonon interaction is most likely responsible for superconductivity in these materials.

We present results of the first optical and angle-resolved photoemission study on a layered ruthenium oxide system with various Ca͞Sr substitution levels. Using two-plane ͑Sr 12x Ca x ͒ 3 Ru 2 O 7 and one-plane Ca 2 RuO 4 crystals we were able to study evolution of the electronic properties in a range from a band metal ͑Sr 3 Ru 2 O 7 ͒ to a "bad" metal ͑Ca 3 Ru 2 O 7 ͒ to a Mott-Hubbard insulator ͑Ca 2 RuO 4 ͒. Apart from a Mott-Hubbard metal-insulator transition (MIT), we have uncovered a qualitative change of the electronic properties at a critical x 0.33. We suggest that the latter is a result of a quantum phase transition into an antiferromagnetic phase that precedes the real Mott-Hubbard MIT. [S0031-9007(98)07196-8]

No abstract

The ab-plane optical spectra of one underdoped and one nearly optimally doped single crystal of HgBa2Ca2Cu3O 8+δ were investigated in the frequency range from 40 to 40,000 cm −1 . The frequency dependent scattering rate was obtained by Kramers Kronig analysis of the reflectance. Both crystals have a scattering rate gap of about 1000 cm −1 which is much larger than the 700 cm −1 gap seen in optical studies of several cuprates with maximum Tc around 93 K. There appears to be a universal scaling between scattering rate gap and maximum Tc for the cuprate superconductors.PACS numbers: 74.25. Gz, 74.72.Gr, 78.20.Ci Although high temperature superconductivity in the copper oxides was discovered over a decade ago, an understanding of the mechanism that gives rise to the high transition temperature, T c , is still elusive. Most of the research has been directed towards the one-and twolayer systems which have a maximum T c in the 93 K range. Recently, however, high quality single crystals of the three-layer Hg based materials with a maximum T c ≈ 135 K have become available [1] making it possible to examine, spectroscopically, oxide superconductors that have significantly higher T c .Of particular interest is the magnitude of the normal state pseudogap. This is a partial suppression of the density of low energy excitations seen well above T c in all underdoped high temperature superconductors. A pseudogap has been seen with a variety of techniques such as angle-resolved photoemission (ARPES), tunneling spectroscopy, specific heat, dc resistivity, nuclear magnetic resonance and optical spectroscopy [2]. Measurements of dc resistivity [3] and NMR 63 Cu T 1 [4] on the threelayer underdoped HgBa 2 Ca 2 Cu 3 O 8+δ (Hg-1223) show evidence of a pseudogap with onset temperatures, T * , of 320 K and 230 K respectively.The size of the pseudogap in the one-and two-layer materials is of the order of 9.5k B T c at optimal doping and essentially independent of temperature. As T c is reduced below optimal doping the pseudogap increases in size [5][6][7].In ab-plane infrared response, the pseudogap is most clearly seen in the frequency-dependent scattering rate which can be calculated from the reflectance after Kramers-Kronig analysis. Here the pseudogap is a suppression of scattering below a characteristic energy which is taken to be a measure of the "size" of the pseudogap.

No abstract