Process optimization of gold stud bump manufacturing using artificial neural networks

摘要

The optimal operating conditions of a gold stud bumping process were determined using a process optimization scheme for a microelectronic packaging foundry. The schematic procedure of the process optimization is first to evaluate effects of the operating parameters on bump size and height, and shear stress, using a design of experimental method. Several operating parameters, such as compression force (bonding load) and electronic flame off (EFO) current and time, were analyzed to affect the formation of the stud bump significantly in the bumping process. Artificial neural networks (ANN) modeling was adopted to establish the relationship between the operating parameters and the bump properties with the experimental data. Some optimization cases of the bumping process with constraints were evaluated using the optimization scheme.