A Full-Process Method from Composition Design to Glass Forming Ability Prediction in Zr-Al-Co-Cu Metallic Glasses

In this paper, a systematic approach utilizing clusters and mixing entropy is proposed to design efficient glass-formers and evaluate glass forming ability (GFA) without relying on thermal properties’ parameters. Firstly, under the guidance of this method, glass-formers in Zr-Al-Co-Cu system were designed as mixtures of topologically packed Zr-Al, Zr-Co and Zr-Cu clusters. Among these, two novel clusters, namely a trigonal prism Co-Zr9 and an Archimedean octahedral anti-prism Co-Co2Zr8 were obtained and used. The best composition is Zr55.79Al12.66Co7.74Cu23.81, expressed as Cu (Cu7+Zr5)+0.867Co (Co2+Zr8)+0.851Al (Al4+Zr8), which exhibits a critical diameter of up to 10 mm. Additionally, based on this method, a novel parameter (ξ), independent of thermal properties’ parameters, was proposed to predict the GFA of metallic glasses. The results indicate that the lower ξ value correlates with better GFA. Simultaneously, the average local five-fold symmetry (ALFFS) from molecular dynamics (MD) simulations was employed to provide supplementary explanations for a few outlier points, further validating the proposed method’s effectiveness. The findings not only confirm the effectiveness of the clusters and mixing entropy approach in designing high-GFA metallic glasses, but also demonstrate the new parameter’s potential in predicting GFA. This study provides important theoretical and practical guidance for optimizing the design of metallic glasses.