The term of defect volatility is introduced as transitions between active and inactive defect states in Subsection 2.1.6. Volatility describes the phenomenon that defects repeatedly disappear (become electrically inactive) and reappear (become electrically active) during measurements. In other words, a volatile defect does not capture or emit charge carriers, schematically shown in Figure 5.13 bottom. In the previous section, the results for the recovery after mixed NBTI/HC stress are presented and it is shown that recovery can be seriously reduced due to non-equilibrium processes. Even source-side defects can remain neutral after mixed NBTI/HC stress as shown schematically in the central panels of Figure 5.13. Although the neutrality of some defects after mixed NBTI/HC stress has the same consequence for recovery as volatile defects, namely no contribution to recovery, there is a fundamental difference between the neutral defects during recovery and volatile defects. Neutral defects affect only the recovery after mixed NBTI/HC stress but not recovery after homogeneous NBTI stress while the volatile defects disappear from all measurements and affect recovery after homogeneous NBTI stress as well. It is mentioned that the volatility of the characterized defects in the previous section was checked regularly by applying homogeneous NBTI conditions and measuring if the observed defects are still electrically active. None of them were volatile during the measurements presented in the previous section. However, following the termination of this study after thousands of cycles of mixed NBTI/HC stress, some defects simply disappeared from the homogeneous NBTI checks. This could be attributed to volatility and is discussed in this section.
In the following is extracted according to Equation 5.1, but with a different lower limit: s s. One main difference to the previous section is, that in this section is the recovery after homogeneous NBTI stress and not recovery in general.
Measurements on large-area devices shown in Figure 5.20 illustrate that the behavior of depends strongly on the “history" of the device. is extracted from the recovery traces after homogenous NBTI stress with −2.5 V, 1 s and 3 ks. In between measurements of , stress and recovery cycles of different stress conditions were performed. For example, the measurement shown in Subfigure 5.20a consists of the following alternating cycles applied subsequently to the same device: measurement of , 1 s NBTI2 stress, 3 ks recovery, measurement of , 1 s NBTI2 stress, 3 ks recovery and so on. In this context, NBTI2 stands for a homogeneous NBTI stress with a different gate bias than the one used for the measurement of , −1.6 V. From the bottom subfigure it can be seen that no considerable reduction of with respect to the cumulative NBTI2 stress time () can be seen. The reduction is less than 0.5 mV in average at 125 °C, which corresponds to approximately 3 %. In this regard, it has been shown recently, that a considerable reduction of the recoverable component after alternating homogeneous NBTI stress is measurable at higher temperatures and longer stress times [127–129]. For example, in [127], shows a reduction of 25 % after 30 cycles of 10 ks stress/10 ks recovery at 200 °C for a similar technology as it is used for the measurements.
By contrast, reduces significantly even at 125 °C if a mixed NBTI/HC stress has been applied previously, as shown in Subfigure 5.20b. The mixed NBTI/HC stress and recovery cycles were performed instead of the NBTI2 cycles. , and of the mixed NBTI/HC stress and recovery were similar or the same as of the NBTI2 cycles but with a −2.5 V applied. The measurement consists of the following alternating cycles applied subsequently to the same device: measurement of , 1 s mixed stress, 3 ks recovery, measurement of , 1 s mixed stress, 3 ks recovery and so on. With respect to the cumulative mixed NBTI/HC stress time () reduces by 2.8 mV, which corresponds to 22 % at 125 °C. A quite similar reduction of due to previously applied mixed NBTI/HC stress at the same temperature has been already observed [72].
It has been introduced previously that can be understood as the cumulative contribution of defects, which have been charged during stress and emit charge carriers during the recovery phase. If is reduced considerably, some of the defects must have disappeared in terms of electrical activity and thus do no longer contribute to . However, from the measurements on large-area devices the volatility of individual defects cannot be observed because their individual contributions to the traces cannot be resolved. Therefore, the volatility checks performed on nano-scale devices (results presented in the previous section) can provide a detailed insight into the behavior of individual defects. The characterized defects in this regard are listed in Table 5.2.
Similar to the measurement sequences shown in Subfigure 5.20b alternating homogeneous NBTI and mixed NBTI/HC stresses were performed on the nano-scale devices. The purpose of applying the homogeneous NBTI stress was comparable to the one in the measurements on the large-area devices, the extraction of . Additionally, the activity of the observed individual defects was checked. In this context, the measurements on device B as a representative device for all measured pMOSFETs are discussed.
Transistor | Def. Nr. | Def. Name | Volatility | ||
A | 1 | A1 | 0.40 | no | |
2 | A2 | 0.21 | no | ||
3 | A3 | – | no | ||
4 | A4 | 0.32 | no | ||
5 | A5 | 0.17 | no | ||
B | 1 | B1 | 0.71 | yes | |
2 | B2 | 0.82 | no | ||
3 | B3 | – | yes | ||
4 | B4 | – | yes | ||
C | 1 | C1 | 0.81 | yes | |
2 | C2 | – | yes | ||
3 | C3 | 0.86 | yes | ||
D | 1 | D1 | 0.20 | no | |
Table 5.2: Defect volatility after different stress conditions: Summary of the observed defect volatility. While the defects B3 and B4 showed a regular volatile behavior by getting inactive and active again from time to time, the defects B1, C1, C2 and C3 were volatile after all stress conditions.
Figure 5.21 is a summary of all recovery measurements after homogeneous NBTI stress of device B. In this device, especially the defects B1 and B2 were monitored. As can be seen, while increases with respect to the cumulative stress time of the mixed NBTI/HC stress cycles 1–4, reduces from 6.5 mV to 5 mV. After all mixed NBTI/HC stress cycles 1–5 (before baking) has reduced to 2 mV. From the mean number of emissions (average over 100 traces of one measurement) and the step heights of the observed defect B1 and B2, the reasons for the reduction becomes clear: on the one hand, the number of emission events reduces, while on the other hand the step heights of the defects change. In this particular case mainly the change of the step height of B1, which is the largest in this device, and the inactivity of B1 before baking contribute to the reduction of . After three days of baking at 280 °C, B1 becomes active again, as shown in cycle 6 in Figure 5.21, with a slightly different step height. During this last cycle, mixed NBTI/HC cycles was applied until the MOSFET failed completely.
Regarding the step height change during the cycles 1–4 in the particular case shown in Figure 5.21, it has to be mentioned that both trends were measured in all devices, increasing and decreasing. In the small sample set, it appears that the decreasing trend was more likely for defects with larger step heights, which dominate . This concerns B1, A1 and C1 listed in Table 5.1. Interestingly, according to the extraction of the lateral position in [104, 105, 124] these are the defects located laterally near the center of the oxide. Whether a defect changes its step height during operation is predominatly due to deviations of the electrostatic surrounding of the defect as discussed in [16]. This means that as soon as other defects are generated/activated or annealed/deactivated in the vicinity of the observed defect, the step height can change. Thus, the reduction of due to step height changes is most probably attributed to activation or deactivation of other defects near the observed defect in the oxide.
The dominant contribution to the reduction of after NBTI stress due to previous mixed NBTI/HC stress is the deactivation of defects, as can be seen in the spectral maps (extracted according to the method described in [40]) in Figure 5.22 for the devices B and C. Especially device C is remarkable in this regard because after 270 ks of mixed NBTI/HC stress all defects observed in this device disappear completely and do not reappear until the device failed completely. With this, reduced permanently to zero. In this context, a few observations on the volatility of defects in the introduced measurements can be summarized.
• Six out of 13 defects show volatility. Three of these six defects, B1, C1 and C3, are located laterally between the center and the drain. For the other three no extraction of their lateral position was possible due to their complicated behavior. While the statistics are small, it appears that it is more likely for defects near the drain to be deactivated with mixed NBTI/HC stress.
• In [58] the probability to find a defect showing volatility in a measurement window of several minutes to one day is estimated with 22 %. In the presented measurements 46 % show volatility in a measurement window up to 100 s, which is considerably more than this previous estimate. If the reaction barrier of the transition between active and inactive defect states is estimated with an Arrhenius law, as proposed in [65], the temperature plays a major role for the transition time constant and thus for the probability that a defect shows volatility or not. Despite the awareness that the estimation in [58] and the percentage found in the experiments are not fully comparable due to different definitions of the experimental window, self-heating effects might be an issue in the context of defect volatility.
• Apparently, there are two types of volatility. The defects B3 and B4, for example, repeatedly dis- and reappear as expected from volatile defects. By contrast, B1, C1, C2 and C3 disappeared after a certain cumulative stress time of mixed NBTI/HC stress and C1, C2 and C3 did not appear again anymore, even after baking and further measurements.
• It has been previously suggested that volatility concerns all oxide defects [65]. However, the continuous reduction of shown in Subfigure 5.20b as well as the permanent deactivation of defects as shown in Subfigure 5.22b is an evidence that some defects are “permanently" annealed. The corresponding transitions to a precursor state has been introduced recently in the context of the permanent component of NBTI degradation [74, 115]. Whether such transitions are truely “permanent" has to be checked in future long-term measurements.