Answers are grounded in this paper and cite their sources.
Measurements of the change in low-temperature thermal conductivity of high-purity single crystal InSb were made upon 2-MeV electron irradiation and annealing. For small defect concentrations, the additive thermal resistivity increases as the 3/4 power of the time-integrated flux Φ rather than the first power as observed for GaAs. The defect thermal resistivity is not proportional to the absolute temperature as predicted by the strain and mass-difference scattering theories of Klemens and Ziman. Irradiation to 5×{}10¹⁷ e/cm² below 52^∘{}K gave 1/K-1K₀=2.7×10^-13Φ3/4 cm-deg/W at 18^∘{}K. A similar result was obtained for measurements at 50^∘{}K. Isochronal anneals with measurements of thermal and electrical conductivity made near 18 and 50^∘{}K indicated almost complete recovery of the thermal conductivity by 395^∘{}K. The increase in thermal resistivity on bombardment at 50^∘{}K is orders of magnitude larger for InSb than for GaAs even though InSb and GaAs exhibit the same increase in lattice strain per 2-MeV e/cm² at 50^∘{}K. It would have been expected that GaAs would be more sensitive to point defects than InSb since GaAs is less limited by umklapp scattering at 50^∘{}K. The strain and mass-difference scattering theories of Klemens and Ziman can not explain the InSb data. The Keyes electron-phonon scattering theory based upon the large effect of strain on the defect electronic wave functions gives a plausible explanation. Annealing begins as low as 40^∘{}K for n-type InSb converted to p type on bombardment, whereas it has been reported that no annealing occurs below 80^∘{}K in unconverted n-type InSb. "Reverse anneals" occur in both thermal and electrical conductivity, implying the existence of at least two competing annealing processes. The activation energy for the anneal of thermal conductivity between 85 and 105^∘{}K agrees with that for the annealing of carrier concentration in p-type InSb, suggesting that the defects measured by electrical and thermal conductivity are the same.
No takes yet. Share an insight, caveat, or question.
F. L. Vook (1964) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: