ABSTRACT
An observer in the laboratory frame sees the change in composition at a point in the sample due to both intrinsic diffusion and the Kirkendall flow of matter. If a diffusion couple is constructed with inert markers at the interface, fixed to the laboratory bench, then the specimen will move relative to the markers during diffusion, a phenomenon known as the Kirkendall effect. On occasions, the concentration dependence of the diffusion coefficients cannot be explained by thermodynamics alone. The sensitivity of the diffusivity of carbon in austenite has been investigated using Eyring's absolute reaction rate theory. The diffusion data for carbon in ferrite have been measured using a variety of techniques including internal friction and mass flow experiments. Carbon atoms in adjacent interstitial sites in austenite or in ferrite repel each other. The diffusion of carbon in tetragonal martensite is therefore expected to be slower than in ferrite.
