Thickness optimization for Intercalation doped Multilayer Graphene Nanoribbon Interconnects

Abstract

In this paper, thickness of Multilayer Graphene Nanoribbon (MLGNR) interconnects is optimized by minimizing the crosstalk delay and noise parameters for intermediate and global level interconnects at 11nm technology node by utilizing the ABCD parameter based model. It is observed that the optimum thickness by minimizing the crosstalk induced delay is 15 nm and 30 nm for perfectly and nearly specular (i.e specularity index, P=1 and 0.8 respectively) intermediate and global level MLGNR interconnects respectively. As far as noise contribution is considered, completely and nearly specular MLGNRs have better immunity to noise when their thickness is less than 15 nm and 175 nm for intermediate and global level respectively. When compared to Cu, fully and nearly specular MLGNRs outperform Cu in terms of delay, noise width and noise area for both intermediate and global level but in terms of peak noise Cu is better than MLGNRs.