A machine learning approach to explore tensile properties of low-temperature solders

Low-temperature solder material has been proposed as a solution in reducing the reflow temperature in order to solve the warpage issue in the advanced electronic packaging. Currently, Sn-Bi solder system is considered as a promising material system for low-temperature solder. Nevertheless, (Bi) phase coarsening after thermal aging causes the brittleness of the solder and thus decreases the reliability in real application. Element doping to Sn-Bi solder is typically used in tailoring its properties. However, tailoring properties in the multi-component system with experimental trial-and-error method is not economically feasible. This study employed machine learning approach to explore tensile properties of low-temperature solders.