Knowledge Bank

Metallic glasses are a relatively new class of engineered material with many unique properties including high yield strength and easy formability. However, they are currently limited by the fact that only limited compositions are known to form metallic glasses, which often contain expensive components. In this work, a computational tool to assess the glass-forming ability of an alloy was developed to help overcome that issue. Through calculating the viscosity of the liquid state and the fraction of icosahedra, we are able to classify alloys by the relative amount of diffusion and driving force for crystallization in the liquid state. The first test of this system on the copper-zirconium binary system identified Cu-35.7at%Zr and Cu-54at%Zr as the best glass-forming alloys, which is consistent with experimental findings. This tool was designed in such a way to be applicable to many different alloy systems and to be simple enough to enable automated testing of many possible compositions.