A PVT Aware Accurate Statistical Logic Library for High-$\kappa$ Metal-Gate Nano-CMOS

Dhruva Ghai1,  Saraju Mohanty1,  Elias Kougianos1,  Priyadarsan Patra2
1University of North Texas, 2Intel Corporation


The semiconductor industry is headed towards a new era of scaling and uncertainty with new key building blocks for the next-generation chips, the high-$\kappa$ metal-gate transistor. There is a need for statistical characterization of high-$\kappa$ metal-gate digital gates as a function of process parameter variations to make them available for designers. In this paper, we present a methodology for PVT aware high-$\kappa$ metal-gate logic library creation while considering the \emph{variability effect in $15$ parameters}. First, statistical models for GIDL current ($\hat{I}{_{GIDL}}$), off-current ($\hat{I}{_{OFF}}$) and drive current ($\hat{I}{_{ON}}$) are presented at the device level. This is followed by statistical characterization of logic cells at room temperature. Data for subthreshold current ($\hat{I}{_{sub}}$), $\hat{I}{_{GIDL}}$, dynamic current ($\hat{I}{_{dyn}}$) and delay is presented. This is followed by results for PVT aware characterization of logic cells. To the best of the authors' knowledge, this is the first research which provides a PVT aware statistical characterization for high-$\kappa$ metal-gate nano-CMOS based logic gates.