Cu is currently being used as the new generation of advanced interconnects. Beneficial additives or dopants to Cu have been sought to improve the electromigration performance of the Cu interconnects primarily through experimental approaches [1]. As a vital alternative, we have established a virtual simulation procedure to screen the potential dopants to Cu by modeling. There are many factors such as film density, stress, stress- voiding, grain boundary defect / diffusion, interface adhesion / defect / diffusion, grain structures (texture, grain size and size distribution) and so on that can affect the electromigration. Here we assume that Cu diffusion along the grain boundaries (GBs) is the dominant mechanism that is responsible for the electromigration performance of the Cu interconnects. As schematically shown in Fig.1, if a dopant is added to Cu, most of the dopant will reside in bulk initially. In order for the dopant to play a beneficial role, it has to be able to segregate to the grain boundary. Then, the dopant is supposed to slow down Cu diffusion along the grain boundary and this can be achieved if the dopant can increase the overall activation energy of Cu grain boundary diffusion.