Single-spin fluid, spin gap, andd-wave pairing inYBa2Cu4

Abstract

We present results of 0 and ' Cu nuclear magnetic resonance (NMR) and nuclear quadrupolar resonance (NQR) studies in the normal and superconducting state of the 82-K superconductor YBazCu408. The various components of the Cu and 0 Knight-shift tensors show strong but similar temperature dependences over the temperature range from 8.5 to 300 K in both the CuOz planes and the chains, supporting the picture that there is only one spin component in the planes and the chains, although with difFerent susceptibilities. The oxygen data obey the Korringa relation. This may be interpreted as Fermi-liquid behavior of the electronic system far away from the antiferromagnetic wave vector. The temperature dependence of both the planar Cu and 0 shift tensors and the planar Cu spin-lattice relaxation rate suggest the opening of a pseudo-spin-gap well above T, The . very different temperature dependence of 1/Ti at the planar 0 and Cu sites points to the reduced role of the antiferromagnetic correlated spin fluctuations at the 0 site. The data favor the conclusion that YBazCu408 is a d-wave superconductor. Evidence is provided by three data sets: the chain Knight shifts, the ratio of the planar copper and oxygen relaxation rates, and the individual low-temperature behavior of these rates.

Anisotropy of the antiferromagnetic spin correlations in the superconducting state of YBa2Cu3O7 and YBa2Cu4O8

“…Great emphasis has been put on the different temperature behaviour of the copper and oxygen relaxation rates in the normal state. However, less attention has been payed on the ratios of these rates [2][3][4][5][6][7][8][9] in the superconducting state. A difficulty arises since in the normal state, the spin-lattice relaxation is largely insensitive to the strength of the applied magnetic field [10,11], while the relaxation rate becomes field dependent in the superconducting state at low temperature, with 1c [5][6][7]12].…”

“…1c data from Nandor et al [16]. The ratio for YBa 2 Cu 4 O 8 (half-filled triangles) was combined in high field for the oxygen measurement only, from the data from Bankay et al [9]). The dashed line is the model prediction for vanishing AFM correlations (K ab 01 = 0).…”

Phenomenological Model of Protected Behavior in the Pseudogap State of Underdoped Cuprate Superconductors

“…We further find (see Fig. 2), from our analysis of the spin-lattice relaxation rate measured in NMR [8], YBa2Cu4O8 [9,10,11], YBa2Cu3O6.63 [12], and Y1−xPrxBa2Cu3O7 (x = 0.05, 0.1, 015, 0.2) [13], compared to the scaling function obtained by Nakano et al [2] (solid line) for the bulk spin susceptibility in La2−xSrxCuO4. T * is the temperature at which the Knight shift has a maximum.…”

Oxygen isotope effect of the plane-copper NQR frequency inYBa2Cu4OMali1, Roos2, Keller3 et al. 2002Phys. Rev. BSelf Cite10060View full textAdd to dashboardCite

“…In an effort to better discern the effect of the new electronic feature at low temperatures we attempt to extrapolate the normal hightemperature behaviour of ν Q (Cu2) to temperatures below 200 K. In doing so we assume that the power law used for ν Q (Ba) holds also for the lattice part of Cu2 EFG. After subtracting this extrapolation from the original ν Q (Cu2) data we get a "rest" whose temperature dependence looks very familiar, namely like the Cu2 Knight shift temperature dependence turned on head [11]. Lacking more insight we decide to fit this "rest" by an empirical function B tanh 2 ( ∆ 2T ), that is very reminiscent of the empirical spin-pseudogap function used to fit Cu2 Knight shift and spin-lattice relaxation data [1,11].…”

“…After subtracting this extrapolation from the original ν Q (Cu2) data we get a "rest" whose temperature dependence looks very familiar, namely like the Cu2 Knight shift temperature dependence turned on head [11]. Lacking more insight we decide to fit this "rest" by an empirical function B tanh 2 ( ∆ 2T ), that is very reminiscent of the empirical spin-pseudogap function used to fit Cu2 Knight shift and spin-lattice relaxation data [1,11]. In the actual ν Q (Cu2) vs. T fit we combine the above two steps and use the expression:…”