Simulative Comparison of Polymer and Ceramic Encapsulation on SiC-MOSFET Power Modules under Thermomechanical Load

Abstract

The need for small, high efficient (98%) and high temperature capable $(200^{\circ}{\rm{C}})$ power electronics lead to the development of SiC-power modules, which can satisfy these advanced requirements. Power electronics have often to resist harsh environments and provide high reliability. A typical way to fulfill these demands is the application of silicone gels and mold compounds, which can be used up to $200^{\circ}{\rm{C}}$. In order to protect power electronics for application at higher temperature up to $250^{\circ}{\rm{C}}$ and overcome the low thermal conductivity of polymer encapsulates new ceramic compounds were developed. Hence they are new to the market, there are less experiences using this compounds to increase reliability of SiC power modules. The aim of this study is to compare the effect of organic mold compounds and ceramic compound encapsulates onto a single chip SiC power module with finite element modeling (FEM) of power cycling loads. With the used assembly and interconnection technologies (AIT) such as active metal brazed (AMB) ceramics, silver sintering and Al|Cu-ribbon bond hybrids, the expected failure is a ribbon bond-lift off. The focus is therefore on the elastic and plastic deformation in the ribbon bond foots. As the stiffness of the encapsulate increases, the strain in the ribbon bonds can be reduced significantly. This allows a significant increase in reliability and lifetime, if the encapsulation does not introduce new failure mechanism. Through a comparison of FEM-results to experimental tested samples, the simulations are validated and a lifetime prediction is made.