摘要

Drain-induced barrier lowering (DIBL) variation is a significant challenge in integrated device and circuit designs for nanomete rmetal-oxide-semiconductor field effect transistors (MOSFETs). This letter analyzed and proposed a compact model to correctly describe DIBL variation induced by random dopants in a 22-nm MOSFET channel considering the second-order effect when using propagation of variation (POV) methodology. The model was validated using a Monte Carlo simulation and captured DIBL's statistical distribution very well. Our model and simulation both revealed a 3 sigma design margin for DIBL due to channel doping is around 31.5 mV/V. The presented statistical modeling approach gives a new perspective for the process-variation-aware design of the nanometer CMOS circuit.