A device-level characterization approach to quantify the impacts of different random variation sources in FinFET technology

A simple device-level characterization approach to quantitatively evaluate the impacts of different random variation sources in FinFETs is proposed. The impacts of random dopant fluctuation are negligible for FinFETs with lightly doped channel, leaving metal gate granularity and line-edge roughness as the two major random variation sources. The variations of Vth induced by these two major categories are theoretically decomposed based on the distinction in physical mechanisms and their influences on different electrical characteristics. The effectiveness of the proposed method is confirmed through both TCAD simulations and experimental results. This letter can provide helpful guidelines for variation-aware technology development