Simulation of diffusion with non-equilibrium vacancies, Kirkendall shift and porosity in single-phase alloys

Abstract

In an alloy subject to vacancy-mediated diffusion, differences in intrinsic diffusivities tend to produce vacancy excess and deficit. These are accommodated by mechanisms such as dislocation climb and pore formation. This is of concern in high temperature alloy-coating systems used in industrial applications, as pores may develop at the interface between the alloy and the coating, which is undesired. This paper presents a multicomponent diffusion model with two types of vacancy sinks/sources: one is associated with dislocation climb and generates lattice shift, the other one is associated with porosity increase/decrease. The model is designed toward a 1D implementation, and porosity is described with a local average volume fraction. Thermodynamic properties and mobility are modeled according to the Calphad method to allow future application to engineering materials. Finite-difference simulations run on two binary systems, NiCr and NiSi, illustrate the role of the two types of sinks in interdiffusion and pore development. Diffusion is found to be more sensitive to the sink strengths in the NiSi system, where intrinsic diffusivities have a stronger composition dependence. This work provides a basis for the evaluation of the parameters involved in vacancy generation/annihilation (e.g. dislocation density) from experimental data, such as concentration profiles obtained from diffusion couple experiments, and for the prediction of porosity in engineering materials.