Towards the systematic study of aging induced dynamic variability in nano-MOSFETs: Adding the missing cycle-to-cycle variation effects into device-to-device variation

In this paper, the NBTI induced dynamic V
_th
variability in nano-scaled MOSFETs is comprehensively studied. In addition to the device-to-device variation (DDV) of NBTI degradation, the non-negligible cycle-to-cycle variation (CCV) due to the random occupation of trap states in each operation cycle is observed for the first time. By using the statistical trap-response (STR) technique, a new characterization scheme of the NBTI induced total ΔV
_th
variation is proposed, combining both DDV (random spatial distribution of trap positions) and CCV together. This systematic study clarifies the superposition of the missing part of trap occupancy probability in the NBTI reliability induced dynamic variation. The single-trap occupancy behaviors as well as the comparison between DC/AC NBTI effects are also studied for microscopic understanding and accurate prediction of the dynamic V
_th
variation due to NBTI degradation in ultra-scaled devices.