Discussion group (DG) summary: NBTI

The negative bias temperature instability (NBTI) is the major reliability issue in advanced CMOS devices. The phenomenon is most readily observed as a negative shift in threshold voltage and degradation in transconductance in pMOS devices which are subject to elevated temperature and inversion stress conditions. NBTI´s importance stems from the fact that despite ∼40 years of research, the details of the phenomenon have still proven quite elusive. The major reason that the governing fundamental physics are still unknown is that NBTI degradation (threshold voltage shift and transconductance degradation) recovers very quickly when the stress is removed. Recovery happens so fast that every single NBTI measurement is subject to a certain degree of error. These errors result in an underestimate of the “true” NBTI degradation and have serious implications for device and circuit lifetime estimates. For the last few decades the NBTI phenomenon has been described using a reaction-diffusion model which links degradation to an interface state generation mechanism involving the depassivation of Si-H bonds at the Si/SiO2 interface and the subsequent diffusion of the hydrogen away from the interface. Recovery is then explained by back diffusion of the hydrogen and subsequent repassivation of silicon dangling bond interface states. Work published from a variety of authors in the last few years has focused on the fact that the reaction-diffusion model has serious deficiencies explaining the experimentally observed recovery trends. This disagreement has fueled great controversy with several authors arguing for a modified reaction diffusion model while others advocate its abandonment.