Development of an elaborate simulation tool for electrochemical failures in microelectronic packages

摘要

The ever increasing complexity and function integration of microelectronic products in combination with the decreasing design margins, the decreasing time-to-market, and ever increasing gap between technology advance and fundamental knowledge opposes a severe challenge for the microelectronics industry to meet the quality, robustness, and reliability requirements of their products. In order to meet these requirements, the reliability of microelectronic products is traditionally assessed using tests at elevated external stimuli, such as temperature, ambient humidity and applied voltage. Recently, the perspective of reliability assessments has shifted towards an approach referred to as knowledge-based qualification, where costumer requirements and operational conditions are translated to stress tests conditions using computer simulations for failure mechanisms and reliability data from corresponding products under comparable conditions. While in the past years simulations tools to predict water absorption and (thermo-)mechanical stresses in packages have been developed, there are no generally accepted simulation tools to predict the effect of electrochemical processes on the performance of products. However, simulation tools that are capable of modelling the electrochemical processes at the interior of packages are indispensable instruments to rigorously study failures due to, e.g., the corrosion of bondpads or the growth of dendritic deposits at metallizations. In this talk a model for the transport of ionic species coupled to a relation for the electrochemical charge transfer rate at electrode is presented. We show results of this model for realistic two-dimensional structures and compare the results with experimental data. We will show that the experimental and model results agree well each other. Additionally, we will show that the model we present can be unequivocally incorporated in the current thermo-mechanical simulation models. Finally, we will address future-n-n trends and discuss the perspectives of elaborate simulation tools for the prediction of microelectronics reliability.