An investigation of the vacancy annealing kinetics and precipitate structure in quenched gold

Abstract

Polycrystalline gold specimens, in the form of narrow foils, with initial nominal purities of 99·999% or greater, were annealed for various periods in air at elevated temperatures, by resistance heating, in order to purify them. The resistivity ratios (ρ°_25°K/ρ°_4·2°K) of the specimens were thus varied in the range from 877 to 5838. These specimens were then rapidly quenched (> 5 × 10^4°c/sec) from either 700°c or 900°c into brine at ∼ 1·7deg;c and 24·5°c, respectively. The electrical resistivity annealing kinetics of the quenched-in vacancy defects were then investigated for annealing temperatures of 40°c and 60°c. Subsequently, the final vacancy precipitate structures in these specimens were observed using transmission electron microscopy. The following results were obtained for the specimens quenched from 700°c and 900°c: (1) The resistivity annealing rate decreased continually for increased specimen purity, over the total range of purity investigated. (2) A large increase in the density, with a corresponding decrease in the size, of the vacancy precipitates observed after annealing at 40°c, and subsequently at 60°c, occurred with increased dissolved impurity content. (3) The measured effective migration energy, E _eff ^m, was found to be independent of specimen purity, over the total range of purity investigated. A value of E _eff ^m=[0.70±0.04] ev was obtained for excess vacancy resistivity increments in the range 0.331 × 10⁻⁹Ω cm≤Δ_ρv≤4.18× 10⁻⁹Ω cm, and for annealing temperatures within the limits 40°c ≤ T _A ≤ 100°c. The observed impurity effects were explained by a model in which small concentrations of certain specific impurities acted as efficient heterogeneous nuclei for the vacancy precipitation. The diffusion distance required for the annihilation of the vacancy defects was thus decreased by an increase in the impurity concentration, with no resulting change in the effective migration energy. Selective impurity doping experiments were also carried out, and it was found that copper impurity atoms were probably responsible for the observed impurity effects.