High-Throughput Determination of Composition-Dependent Interdiffusivity Matrices and Atomic Mobilities in fcc Cu-Ni-Al Alloys by Combining Diffusion Couple Experiments with HitDIC Modeling

Abstract

In this paper, eight solid-state diffusion couples in fcc ternary Cu-Ni-Al alloys at 1073, 1173 and 1273 K were prepared, and the corresponding composition profiles were measured with an electron probe microanalyzer (EPMA) technique. Based on the measured composition profiles and HitDIC software, a set of reliable atomic mobility descriptions in the fcc Cu-Ni-Al system were directly established, from which the composition- and temperature-dependent interdiffusion coefficient matrices were evaluated. Then, the three-dimensional (3D) main interdiffusivity surfaces for fcc Cu-Ni-Al systems were constructed. The diffusion rate of Al is faster than that of Ni. Both the main interdiffusivities of Al and Ni increase with the increase of Al concentration but decrease with Ni concentration. The reliability of the evaluated interdiffusivities was first validated by reproducing the measured composition profiles and interdiffusion fluxes. After that, the interdiffusivities and corresponding uncertainties at intersection points of diffusion couples were evaluated by using the Matano–Kirkaldy method combined with distribution functions. Especially all the interdiffusion coefficients estimated from the present mobility parameters are within the bounds of the uncertainties, which further quantitatively verified the reliability of the present mobility descriptions. The accuracy of the as-determined mobility is further highlighted via combining with the reported thermodynamic descriptions, which enables an accurate prediction of the volume of Ni3Al precipitates in the Cu-Ni-Al alloy during artificial aging. The present work indicates that with the accurate thermodynamic as well as boundary mobility descriptions, reliable ternary mobility descriptions can be achieved by combining the diffusion couple experiments with the HitDIC modeling.