Research question
How do material mismatch, layered geometry, and thermal loading interact to produce package warpage and interfacial stress—and which model inputs explain the discrepancy between simulation and test?
Approach
I build and run implicit thermo-mechanical finite element models of layered, multi-material chip packages in Abaqus. The work combines mechanics-based parameter studies with full-field experimental measurements, so model credibility is judged by spatial response rather than a single scalar quantity.
- Model temperature-dependent properties and coefficient-of-thermal-expansion mismatch across bonded layers
- Study geometry, layer thickness, material selection, and thermal load cases parametrically
- Rank the dominant contributors to warpage and interfacial stress
- Compare predicted and measured fields, then use the residuals to calibrate model inputs
Why it matters
Advanced packages concentrate multiple materials, interfaces, and length scales into a tightly coupled structure. Reliable design requires more than a fast simulation: it requires a model whose assumptions and parameters are tested against the actual package response.
This is SRC-sponsored research at UT Austin.