Initiation and propagation of delaminations at the underfill/passivation interface in flip-chip assemblies

The International Technology Roadmap for Semiconductors continues to highlight the need for a better understanding of interfacial behavior and a heightened ability to characterize and control interfacial strength, as areas crucial to future chip development and manufacturing. Interfacial fracture mechanics approaches based on geometries such as double cantilever beam and four-point bend have been widely accepted as means to quantify interfacial adhesion in terms of resistance to propagation of existing delaminations, however difficulties arise in rigorously applying this methodology to small-scale delamination and disbond initiation problems. A promising "fracture mechanics-like" alternative instead looks at edges and corners as singular initiation sites for delamination and can be used to predict failure in the absence of a preexisting disbond. Initiation is predicted to occur when the stress intensity factor of the singularity reaches a critical value, similar to interfacial fracture mechanics where propagation occurs when the stress intensity factor of the crack tip singularity becomes critical. This work is aimed at understanding how the initiation and propagation phenomenon relate and the interplay between the molecular and mechanical contributions differ between the two events.