« PreviousUpNext »Contents
Previous: 7.2 Interaction of hydrogen with degradation and recovery dynamics    Top: 7 The role of hydrogen in NBTI degradation    Next: 8 Summary and Conclusions

7.3 Conclusions on the role of hydrogen in the NBTI

By varying the BEOL processing (layer sequence – Section 7.1; Ti barrier thickness – Section 7.2), different split wafers were produced which provide vastly different hydrogen contents within the gate oxide. The initial variance in the hydrogen budget has been evaluated electrically by means of CP measurements and physically by means of TOFSIMS analysis. After subjecting the individual samples to NBTS, the degradation and recovery signatures have been investigated by making use of the newly developed characterization techniques. In agreement with our proposed microscopic model, it has been demonstrated that the recoverable part of NBTI degradation is largely independent of hydrogen while the quasi-permanent (math image) shift component is strongly linked to the total hydrogen budget within the gate oxide. Also, a strong correlation between the increase in the CP signal and the quasi-permanent (math image) shift component was found which is also fully consistent with our model.

« PreviousUpNext »Contents
Previous: 7.2 Interaction of hydrogen with degradation and recovery dynamics    Top: 7 The role of hydrogen in NBTI degradation    Next: 8 Summary and Conclusions