Chapter 9
Modeling NBTI in High-k SiGe pMOSFETs

In the last chapter it was concluded that the recovery after BTI is the mere consequence of single defects being discharged at certain emission times, which gives a step-like drain current behavior in small devices. The superposition of many of these defects, as observed in large-area devices, yields the typical log-like recovery behavior[11]. The latest attempt to model such defects is based on the non-radiative multi-phonon emission (NMP) theory after [153125124], cf. Chapter 8.5.3. This theory assumes the conservation of the total energy of a defect or defect system consisting of a strongly coupled electronic and vibronic part [130].

In [111] the NMP model was already shown to successfully reproduce measurement data of small-area devices containing only a few defects. In this chapter it will be shown that the theory also holds for rather complex large-area p-MOSFETs containing a larger number of defects. Such devices have been studied by Franco et al. [164165] and feature a buried SiGe channel with a thick SiGe quantum well of high Ge-fraction (55%  ) and a thin silicon cap below the high-k dielectric in order to reduce NBTI. This type of device is schematically depicted in Fig. 9.1. Devices of this kind were subjected to NBTI stress using various stress voltages and temperatures via the extended measure-stress-measure routine after [18]. For this, a static ID(VG )  -characteristic is taken first to obtain a reference. After the stress sequences with logarithmically increasing stress times from 2s  to 2000s  the degraded threshold voltage is monitored with a delay of 2ms  .


PIC


Figure 9.1: Schematic view of the high-k gate-stack device including a thin SiCap and a high-mobility SiGe-layer as quantum well in the channel region to reduce NBTI [165].


 9.1 Inverse Modeling
 9.2 Multi-State Defect Model
  9.2.1 Distribution of Defects
  9.2.2 Reservoir of Holes - Classical vs. Quantum Mechanical Description
 9.3 Results
 9.4 Conclusions