Computational Materials Science and Engineering Education: An Updated Survey of Trends and Needs

We present a summary of the state of computational materials science and engineering (CMSE) education based on a survey of materials science department chairs, faculty with computational interests, and employers of materials scientists and engineers. This survey is an update of one previously conducted in Thornton et al. (JOM 61:12–17, 2009). Three questionnaires were distributed among department chairs, faculty, and employers. The surveys asked to rate the importance of incorporating CMSE into the undergraduate curriculum, how it should be incorporated, the current offerings in CMSE, how those offerings have recently changed, what software tools are taught/used, and what opportunities exist in CMSE education, along with freeform questions regarding experience in teaching CMSE and impact of CMSE education. The survey results revealed an increased availability of CMSE courses in most of the materials departments surveyed, and strong support for including CMSE into the core curriculum. They also indicated that there is no clear preference as to whether such incorporation into the core curriculum should be through standalone courses or through modules in existing courses. The resources used in these courses, including online tools and software, are also summarized. The responses from the computational faculty point to a continued need for modules, including software tools and educational materials, that can be readily implemented by materials faculty regardless of their area of expertise.     

Continuum simulations of the formation of Kirkendall-effect-induced hollow cylinders in a binary substitutional alloy

We examine binary substitutional diffusion in cylindrical diffusion couples in which free surfaces are considered explicit vacancy sources and sinks. The central region of the cylinder is initially occupied by an atomic species with a larger hop frequency, while the outer region is occupied by another atomic species with a smaller hop frequency. Equilibrium vacancy concentration is maintained at free surfaces that serve as vacancy sources and sinks. In the crystal, diffusion is governed by the standard diffusion equations with analytically evaluated diffusion coefficients. The void growth dynamics and hollow cylinder formation stemming from the Kirkendall effect are simulated. Our results show that the Kirkendall void growth involves two competing factors. One is the net inward vacancy flux that favors void growth. The other is the Gibbs–Thomson effect that favors void shrinkage. We compute the critical initial radius for void growth above which the Kirkendall effect dominates over the Gibbs–Thomson effect. The fully grown void radius and the elapsed time to the fully grown size are also predicted for different fast-diffuser volume fractions and fast-to-slow diffuser atomic hop frequency ratios.     

Detection of thermally grown oxides in thermal barrier coatings by nondestructive evaluation

The thermal-barrier coatings (TBC) sprayed on hot-section components of aircraft turbine engines commonly consist of a partially stabilized zirconia top-coat and an intermediate bond-coat applied on the metallic substrate. The bond-coat is made of an aluminide alloy that at high engine temperatures forms thermally grown oxides (TGO). Although formation of a thin layer of aluminum oxide at the interface between the ceramic top-coat and the bond-coat has the beneficial effect of protecting the metallic substrate from hot gases, oxide formation at splat boundaries or pores within the bond-coat is a source of weakness. In this study, plasma-sprayed TBC specimens are manufactured from two types of bond-coat powders and exposed to elevated temperatures to form oxides at the ceramic-bond-coat boundary and within the bond-coat. The specimens are then tested using nondestructive evaluation (NDE) and destructive metallography and compared with the as-manufactured samples. The objective is to determine if NDE can identify the oxidation within the bond-coat and give indication of its severity. While ultrasonic testing can provide some indication of the degree of bond-coat oxidation, the eddy current (EC) technique clearly identifies severe oxide formation within the bond-coat. Imaging of the EC signals as the function of probe location provides information on the spatial variations in the degree of oxidation, and thereby identifies which components or areas are prone to premature damage.