In this work, we unequivocally demonstrate that strong temperature dependen
ce of time(charge)-to-breakdown T-BD/Q(BD) observed on ultra-thin oxides is
not a thickness effect but rather a consequence of two experimental facts:
(1) voltage-dependent voltage acceleration and (2) temperature-independent
voltage acceleration within a fixed T-BD window. By extending down to -30
degreesC, we found non-Arrhenius temperature dependence is a totally indepe
ndent effect. Based on our statistically accurate experimental database, we
found that defect-generation rate and critical defect density as commonly
measured using stress-induced leak-age current (SILC) only show a change of
two orders of magnitude and no change, respectively. Weibull slopes are al
so found to be temperature-independent. We propose an alternative model of
two-step hydrogen degradation to explain these experimental results. (C) 20
01 Published by Elsevier Science B.V.