Statistical modeling of corrosion-induced wirebond failure

This paper describes the initial results of one portion of a project to develop effective analytical tools for predicting the effect of atmospheric corrosion on the reliability of electronic devices. The specific objectives of this work were to experimentally characterize the atmospheric corrosion of aluminum-gold wirebonds and to develop a statistical-based model that describes the effect of the resulting stochastic process on the reliability of a selected electronic assembly. The experimental characterization included an attempt at accelerated aging. Modeling involved: (1) the development and validation of empirical models that describe the effects of environmental parameters on corrosion rate, and (2) the formulation and validation of a reliability-prediction model using the accelerated aging data and long-term field information as it becomes available. A preliminary assessment of the effect of three environmental factors on wirebond failure rate was performed and an empirical rate model defined. Subsequently, a statistical treatment of the rate information was used in a Monte Carlo simulation technique to determine the service life of a hypothetical electronic assembly. This work demonstrated that stochastic, corrosion-induced degradation can be successfully incorporated in classical techniques to analyze component reliability. 19 figs., 3 tabs