Proceedings of the Royal Society of London A Mathematical and Physical Sciences · 1956 · 81 citations · 13 references
Natural SciencesChemistryMaterials ScienceHeat TransferMaterials PhysicsPhysical ChemistryElectrical ConductivityHeat ConductionHeat Transfer In SolidsIonic ConductivityMetalsThermal Properties Of MaterialsImpurity TransportLattice Thermal ConductivityLow-temperature Thermal PropertiesResistivityTemperature Dependence Of Electrical PropertiesTemperature-dependent ResistivityThermal Conductivity MeasurementsThermal Resistivity
Abstract Measurements of the thermal and electrical conductivities of very pure lithium, sodium, potassium, rubidium and caesium have been made down to temperatures as low as 2°K. The respective resistivities, W and ρ, may be written as the sum of an impurity resistance (W0, ρ0) and a so-called ‘ideal’ component (Wi, ρi) due to scattering by the thermal vibrations of the lattice. The terms in the thermal resistivity may be represented by W0=A/TandW1=BTnforT≤θ/10, where n ≃ 2 and A = (ρ0/2⋅45) x 108 cm deg. 2W-1. Current theory predicts thatt he quantity C ≡ Bθ2/W∞N⅔ should be constant, where N is the number of free electrons per atom and W is the measured high-temperature resistivity. Taking N = 1, the present experiments yield C ≃ 18 ± 4. The electrical resistance may be written ρ=ρ0+βTmforT<θ/10 with m ≃ 5 except for sodium, where, below 8°K , m is found to increase to 6. The theoretical relationships which exist between the low-temperature ‘ideal’ resistivities and those at higher temperatures are discussed in conjunction with the measured values. It is concluded that with the existing theories, no common adjustment of θ can give satisfactory agreement of theory with experiment. A new simple semi-empirical expression is put forward for Wi which provides rather good agreement with experiment.
13