Answers are grounded in this paper and cite their sources.
In Li, the very low diffusion thermopower parameter ξ~-8 is indicative of strong scattering in the p wave, and an anomalously low result is to be expected for any method in which the scattering in one angular momentum component is stronger than that in any of the others. By contrast, when no one component dominates we obtain a substantially positive result (examples being Na, K, and Rb, for which ξ is in the neighborhood of +3). The depressed value for Cs (0.2 in the solid, - 1.3 in the liquid) can be understood in terms of substantial (though, in the above sense, not quite dominant) scattering in the d wave. A standard phase-shift analysis of the electron-impurity scattering explains semiquantitatively the diffusion thermopowers of dilute alkali-alkali alloys, provided energy-dependent scattering data are used. Virtual recoil could provide desirable corrections.
No takes yet. Share an insight, caveat, or question.
Meyer et al. (1969) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: