We report microwave surface resistance (${\mathit{R}}_{\mathit{s}}$) measurements on two very-high-quality ${\mathrm{YBa}}_2$${\mathrm{Cu}}_3$${\mathrm{O}}_{6.95}$ crystals which exhibit extremely low residual loss at 1.2 K (2-6 μΩ at 2 GHz), a broad, reproducible peak at around 38 K, and a rapid increase in loss, by 4 orders of magnitude, between 80 and 93 K. These data provide one ingredient in the determination of the temperature dependence of the real part of the microwave conductivity, ${\mathrm{\sigma }}_1$(T), and of the quasiparticle scattering time. The other necessary ingredient is an accurate knowledge of the magnitude and temperature dependence of the London penetration depth, λ(T). This is derived from published data, from microwave data of Anlage, Langley, and co-workers and from, high-quality μSR data. We infer, from a careful analysis of all available data, that ${\lambda }^2$(0)/${\lambda }^2$(T) is well approximated by the simple function 1-${\mathit{t}}^2$, where t=T/${\mathit{T}}_{\mathit{c}}$, and that the low-temperature data are incompatible with the existence of an s-wave, BCS-like gap. Combining the ${\mathit{R}}_{\mathit{s}}$ and λ(T) data, we find that ${\mathrm{\sigma }}_1$(T), has a broad peak around 32 K with a value about 20 times that at ${\mathit{T}}_{\mathit{c}}$. Using a generalized two-fluid model, we extract the temperature dependence of the quasiparticle scattering rate which follows an exponential law, exp(T/${\mathit{T}}_0$), where ${\mathit{T}}_0$≊12 K, for T between 15 and 84 K. Such a temperature dependence has previously been observed in measurements of the nuclear spin-lattice relaxation rate. Both the uncertainties in our analysis and the implications for the mechanism of high-temperature superconductivity are discussed.

• Received 9 April 1992

DOI:https://doi.org/10.1103/PhysRevB.47.11314