TY - THES
T1 - Numerical Analysis and Experimental Verification of Stresses Building up in Microelectronics Packaging
AU - Rezaie Adli, Ali
PY - 2017
Y1 - 2017
N2 - This thesis comprises a thorough study of the microelectronics packaging process by means of various experimental and numerical methods to estimate the process induced residual stresses. The main objective of the packaging is to encapsulate the die, interconnections and the other exposed internal components by providing mechanical protection, heat dissipation, electrical insulation and etc. It is a three stage process comprising encapsulation of the die, complete polymerization at a preset mold temperature and cooling to room temperature. Thermosetting polymers are used as the encapsulant in this process due to their unique mechanical, thermal and electrical properties. The packaging results in residual stress build-up both during the molding and later due to the cyclic thermo-mechanical loading of the electronic or electromechanical devices in which the encapsulated package is fixed. These residual stresses are initiated by the crosslinking (curing) of the epoxy polymer during the molding. Crosslink formation is the property of the thermosetting polymers, which is accompanied by stiffness build-up and shrinkage under constrained boundary conditions. Besides, the encapsulation molding is conducted at a high cure temperature (≈175oC). Hence, the subsequent cooling to room temperature leads to further shrinkage of the cured polymer along with the other encapsulated package components and the CTE mismatch between the layers adds up to the total residual stress inside the package. The key to a reliable simulation lies in an accurate representation of the material and mechanical behavior of the polymer. In this thesis, the time, temperature and conversion dependent behavior of the epoxy molding compound (EMC) is determined by various experimental methods including DSC, DMA, rheometer and PVT and the relevant material behaviors are modeled and implemented in 1D and 2D numerical methods. For verification of the numerical results a novel experimental method is used providing the real time stress measuring capability during packaging.
AB - This thesis comprises a thorough study of the microelectronics packaging process by means of various experimental and numerical methods to estimate the process induced residual stresses. The main objective of the packaging is to encapsulate the die, interconnections and the other exposed internal components by providing mechanical protection, heat dissipation, electrical insulation and etc. It is a three stage process comprising encapsulation of the die, complete polymerization at a preset mold temperature and cooling to room temperature. Thermosetting polymers are used as the encapsulant in this process due to their unique mechanical, thermal and electrical properties. The packaging results in residual stress build-up both during the molding and later due to the cyclic thermo-mechanical loading of the electronic or electromechanical devices in which the encapsulated package is fixed. These residual stresses are initiated by the crosslinking (curing) of the epoxy polymer during the molding. Crosslink formation is the property of the thermosetting polymers, which is accompanied by stiffness build-up and shrinkage under constrained boundary conditions. Besides, the encapsulation molding is conducted at a high cure temperature (≈175oC). Hence, the subsequent cooling to room temperature leads to further shrinkage of the cured polymer along with the other encapsulated package components and the CTE mismatch between the layers adds up to the total residual stress inside the package. The key to a reliable simulation lies in an accurate representation of the material and mechanical behavior of the polymer. In this thesis, the time, temperature and conversion dependent behavior of the epoxy molding compound (EMC) is determined by various experimental methods including DSC, DMA, rheometer and PVT and the relevant material behaviors are modeled and implemented in 1D and 2D numerical methods. For verification of the numerical results a novel experimental method is used providing the real time stress measuring capability during packaging.
UR - http://resolver.tudelft.nl/uuid:fc197d29-f9bd-4b0e-a4d5-485344d8d429
U2 - 10.4233/uuid:fc197d29-f9bd-4b0e-a4d5-485344d8d429
DO - 10.4233/uuid:fc197d29-f9bd-4b0e-a4d5-485344d8d429
M3 - Dissertation (TU Delft)
SN - 978-94-6186-777-3
ER -