CONNECTING PARAMETRIC AGING TO CATASTROPHIC FAILURE THROUGH THERMODYNAMICS

There have been numerous studies with Arrhenius theory in reliability. However, very few actually apply to parametric reliability analysis. This paper provides fundamental details in this area of reliability ph ysics, We derive a Thermally Activated Time-dependent (TAT) model for both parametric & catastrophic Arrhenius aging. It is shown how aging dynamics depend upon thermodynamics specific to device reliability phy sics and how catastrophic phenomena can be correlated to device life d ynamics. We first demonstrate that Arrhenius degradation behavior can lead to log(time) aging kinetics with lognormal failure times. We disc uss how this helps to explain why the lognormal distribution has been so successfully applied in semiconductor reliability analysis. Log(tim e) aging is extremely important in reliability physics, because its or igin can mathematically be tied to Arrhenius mechanisms, of which nume rous examples exist. Furthermore, this paper ties the reliability phys ics of the thermodynamic process to the log(time) aging expression. As such, the parametric dependencies of the aging process can be found b y analysis of the thermodynamic activated free energy. The appendix pr ovides examples to aid the reader. Next, this work is extended to the catastrophic case. It is shown that parametric aging can mathematicall y be related to catastrophic phenomena for severe degradation due to A rrhenius mechanisms. Linking the two phenomena in this way yields a pr oactive approach to reliability in which catastrophic phenomena are co rrelated to device life. The results demonstrate that early life degra dation trends provide important links to eventual catastrophic phenome na.