Diffusion of Mercury in Tin

Abstract

The diffusion rate of mercury was measured in the tin single crystals along both the $a$ and $c$ axes and is described by diffusion constants: $D_0^a={30}_{-12\mathrm{}}^{+20\mathrm{}}$ ${\mathrm{cm}}^2$/sec, $D_0^c={7.5}_{-3.5\mathrm{}}^{+6.4\mathrm{}}$ ${\mathrm{cm}}^2$/sec, $Q_D^a=26.8\mathrm{}\pm 0.5$ kcal/mole, $Q_D^c=25.3\mathrm{}\pm 0.6$ kcal/mole. This behavior does not differ greatly from the tin self-diffusion and in the light of present understanding of diffusion in polyvalent hosts, it seems most likely that the mercury is dissolved substitutionally and is diffusing by a vacancy mechanism. No good theoretical justification could be given to explain why mercury should prefer substitutional sites in tin, but be at least partially stable in interstitial-vacancy pairs in lead, although the possibility of van der Walls forces is discussed. Mercury diffusion is compared to diffusion of other impurities in tin and a new mechanism is suggested to explain the large anisotropy of diffusion of gold and silver between the tin $a$ and $c$ directions.