Inverse design of Bulk metallic glasses with reduced glass transform temperatures via Target-Optimal inverse projection

Bulk metallic glasses (BMGs) have attracted significant attention due to their unique properties, such as high strength, excellent corrosion resistance, and good wear resistance. However, the discovery of new BMG materials with higher performance remains a challenge. In this study, we present an efficient method for rapidly discovering new BMG materials with higher reduced glass transition temperature (Trg). By combining pattern recognition projection and target-optimal inverse projection (TOIP) methods, we have developed a highly effective approach for identifying BMG materials with improved properties. The identified BMG material from TOIP was verified by machine learning model and experiments. The predicted Trg was 0.6584 and the experimental value was found to be even higher of 0.6658, which surpassed the highest one reported in previous experiments. The result demonstrates the significant potential and cost-effectiveness of the TOIP method for discovering new BMG materials with improved properties. The method proposed in this study can be extended to materials design and controllable synthesis of other material systems.