Experimental constraints on models of normal-state transport

Abstract

Important features of electrical transport in cuprate superconductors are discussed with emphasis on the observations of two distinct dissipative processes in the normal state. The significance of the Matthiessen's rule behavior of both scattering rates is highlighted. Recent evidence that the thermoelectric power is governed by both relaxation processes indicates that the second scattering rate is not related to cyclotron motion as previously conjectured. Transport measurements in the normal state of high T<SUB>c</SUB> cuprate superconductors present a host of tantalizing anomalies. Most experimental probes of electrical and thermal transport properties, including the resistivity, Hall effect, and the thermoelectric power, show intriguing deviations from the behavior expected from a usual Fermi liquid metal. What makes the problem of normal state transport so compelling is that the peculiar features of these transport properties -- their deviations from Fermi liquid behavior -- all show simple systematics. The difficulty is that when each transport property is studied in isolation, its systematics are obvious, but there is no obvious relation between different experiments. What is needed is a unifying picture -- a single transport phenomenology which can reproduce such seemingly unrelated mysteries as the temperature dependence of the Hall effect and the two-component thermopower. The goal of this article is to discuss a number of experimental guides and limitations for the development of a normal-state transport phenomenology. Some features of experimental data, such as the observation of Matthiessen's rule are so simple that they are commonly overlooked. Others, such as the appearance of two distinct lifetimes in the thermoelectric power are quite new and have important consequences. It seems useful to assemble them here.