Search

NMR data from high- ${T}_{c}$ superconducting cuprates having widely different ${T}_{c,\mathrm{max}}$ values and different hole concentrations, $p$, are shown to implicate a $d$-wave-like pseudogap. The deduced pseudogap energies agree well with the gap in the charge excitations determined from photoemission measurements, and the gap energies scale with ${T}_{c,\mathrm{max}}$. The pseudogap and superconducting gap thus have the same energy scale and for different cuprates have a universal $p$ dependence.

Resistance, thermopower, and $^{89}\mathrm{Y}$ nuclear magnetic resonance (NMR) measurements have been carried out on ${\mathrm{YBa}}_{2}$(${\mathrm{Cu}}_{1\mathrm{\ensuremath{-}}\mathit{x}}$${\mathrm{Ni}}_{\mathit{x}}$${)}_{4}$${\mathrm{O}}_{8}$ superconductors. Nickel substitution reduces both ${\mathit{T}}_{\mathit{c}}$ and the pseudogap energy at almost the same rate as does Zn substitution, but unlike Zn substitution, there is no second peak in the $^{89}\mathrm{Y}$ NMR spectra that can be attributed to Y atoms sited near a local moment. We show that this discrepancy is consistent with there being no local suppression of the normal-state pseudogap about the Ni impurity contrary to the case for Zn. \textcopyright{} 1996 The American Physical Society.

We have performed ${}^{63}\mathrm{Cu}$ NMR measurements on the electron-doped high-temperature superconducting cuprate (HTSC) ${\mathrm{Pr}}_{2\ensuremath{-}x}{\mathrm{Ce}}_{x}{\mathrm{CuO}}_{4}$ $(x=0.10,$ 0.15, and 0.20), at 9 T, which is sufficient to suppress ${T}_{c}$ to zero. The Cu spin-lattice relaxation rate and the Cu spin-lattice relaxation rate anisotropy can be consistently interpreted in terms of coupling to antiferromagnetic spin fluctuations for temperatures as low as 6 K. We find that the spin-fluctuation spectrum probed by the Cu spin-lattice relaxation rate does not change with increasing electronic concentration, contrary to a recent theoretical predication. There is no evidence in the Cu spin-lattice relaxation rate data for a temperature-independent spin gap that is as large as theoretically predicted or as large as the normal-state pseudogap energy reported from infrared reflectance measurements on the electron-doped HTSC ${\mathrm{Nd}}_{2\ensuremath{-}x}{\mathrm{Ce}}_{x}{\mathrm{CuO}}_{4}.$ The Cu nuclear quadrupole resonance frequency is significantly smaller than that observed in the hole doped HTSC, which implies a nearly complete cancellation of the Cu $3d,$ O $2p,$ and nuclei contributions to the electric-field gradient at the Cu nucleus. The orbital shift anisotropy is similar to that observed in the hole doped HTSC, implying a similar relative splitting of the Cu $3d$ orbitals.

We report the results from a thermopower and a Cu nuclear magnetic resonance (NMR) study of the infinite-${\mathrm{CuO}}_{2}$-layer electron-doped high-temperature superconducting cuprate (HTSC) ${\mathrm{Sr}}_{0.9}{\mathrm{La}}_{0.1}{\mathrm{CuO}}_{2}.$ We find that the temperature dependence of the thermopower, $S(T),$ is different from that observed in the hole-doped HTSC. In particular, we find that $\mathrm{dS}(T)/dT$ is positive above \ensuremath{\sim}120 K. However, we show that $S(T)$ can still be described by the same model developed for the hole-doped HTSC and hence $S(T)$ is not anomalous and does not imply phonon-mediated pairing as has previously been suggested. The Cu NMR data reveal a Knight shift and spin lattice relaxation rate below ${T}_{c}$ that are inconsistent with isotropic s-wave pairing. The Cu spin lattice relaxation rate in the normal state, however, is Curie-Weiss like and is comparable to that of the optimally and overdoped hole-doped HTSC ${\mathrm{La}}_{2\ensuremath{-}x}{\mathrm{Sr}}_{x}{\mathrm{CuO}}_{4}.$ The magnitude of the Knight shift indicates that the density of states at the Fermi level is anomalously small when compared with the hole-doped HTSC with the same ${T}_{c},$ indicating that the size of ${N(E}_{f})$ is of little importance in the HTSC. We find no evidence of the normal state pseudogap that is observed in the hole-doped HTSC and which was recently reported to exist in the electron-doped HTSC ${\mathrm{Nd}}_{1.85}{\mathrm{Ce}}_{0.15}{\mathrm{CuO}}_{4}$ from infrared reflectance measurements.

$^{133}\mathrm{Cs}$ NMR results are reported for liquid Cs-Au alloys ranging from pure Cs to 18% excess Au in the ionic liquid compound CsAu. As one approaches CsAu from the Cs-rich side, a rapid drop in the Knight shift is observed, accompanied by a sharp peak in the nuclear relaxation rate at about 5% excess Cs in CsAu. Analysis of the static and dynamic hyperfine fields shows that the conduction electrons abruptly localize, probably as $F$- center analogs, when the concentration of excess Cs falls below about 7%.

Measurements of the $^{23}\mathrm{Na}$ nuclear-relaxation time ${T}_{1}$ are reported for two types of Na $\ensuremath{\beta}$-alumina specimen, i.e., flux-grown crystals from our laboratory and melt-grown crystals from Union Carbide Corp. A phenomenological calculation of ${T}_{1}$ due to activated hopping motion, taking account of local variations in energy-barrier heights, is presented. This theory is used to interpret our own ${T}_{1}$ data, plus data on ${T}_{1}$ in the literature for specimens prepared by a third method, in terms of barrier-height distributions. It gives a good account of both the temperature and frequency variation of these three contrasting sets of ${T}_{1}$ data, where the differences are attributed to variations in the structure and/or arrangement of charge defects. The characteristics of the barrier-height distributions obtained are discussed in terms of preparation methods and other available data. The rather larger conductivity for melt-grown (as opposed to flux-grown) material reported by Barker et al. is found to be associated with a displacement of activation energies to lower values. Values of the quadrupolar coupling constants from the ${T}_{1}$ data fits as well as from second-order quadrupolar broadening are found to be essentially independent of sample source, suggesting that this coupling is determined primarily by the host structure. Finally, the ${T}_{1}$ fits yield values of the local vibration or "attempt" frequency ${\ensuremath{\nu}}_{0}$ for activated hopping which are \ensuremath{\sim} ${10}^{11}$ Hz for all three cases considered in sharp contrast with the vibration frequency (\ensuremath{\sim} 2\ifmmode\times\else\texttimes\fi{}${10}^{12}$ Hz) reported from infrared spectra and Raman scattering. Our value, which is derived directly from the hopping motion itself, does not yield the good agreement with experiment noted earlier when used in available theories of the diffusion constant.

A $^{59}\mathrm{Co}$ NMR investigation of liquid gold-cobalt alloys has been carried out over the complete composition range. At all compositions, the NMR properties are dominated by orbital effects. The orbital susceptibility increases from 8% of the total susceptibility (at 1700 K) for pure cobalt to about 40% for the dilute alloys. The values for the $d$-spin hyperfine field and the moment lifetime deduced from the data show that the magnetic character of a cobalt atom changes markedly on adding 25 at.% gold to pure cobalt and that the properties at this composition are closer to those of the dilute alloys than of pure cobalt. The $d$-spin hyperfine field decreases by more than a factor of 3 for a 75 at.% cobalt alloy with a small decrease on adding further gold. The nuclear relaxation rates are determined by the fluctuating orbital susceptibility. Calculation of the composition dependence of the rate using an interacting impurity model yields excellent agreement with experiment over the full composition range with the exception of pure cobalt. In this model, the local-moment lifetime is independent of concentration, and the fluctuation time for the local hyperfine field is shortened by exchange coupling between the cobalt atoms. For pure cobalt, we deduce that the moment lifetime is reduced by a factor of 2 to about 5 \ifmmode\times\else\texttimes\fi{} ${10}^{\ensuremath{-}15}$ sec.

The magnetic susceptibilities of the noble metals have been accurately measured close to the melting point in both liquid and solid states. In the liquid the susceptibilities are (in cgs volume units) Cu: (-0.70\ifmmode\pm\else\textpm\fi{}0.01)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$, Ag:(-2.05\ifmmode\pm\else\textpm\fi{}0.01)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$, Au: (-2.74\ifmmode\pm\else\textpm\fi{}0.01)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$, and in the solid Cu: (-0.59\ifmmode\pm\else\textpm\fi{}0.01)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$, Ag: (-1.84\ifmmode\pm\else\textpm\fi{}0.01)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$, Au:(-2.63\ifmmode\pm\else\textpm\fi{}0.01)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$. The results obtained agree well with the susceptibilities calculated recently by Borchi and de Gennaro if the effects of electron-electron interactions are correctly taken into acount.