Here is the abstract you requested from the MMC_2010 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Challenges in Assembly and Reliability of Package on Package|
|Keywords: Package-on-Package , Assembly, Challenges|
|Package on Package (PoP) was developed to integrate the logic and memory devices primarily found in portable consumer products. Stacking the devices allows for vertical expansion while minimizing overall footprint. Several levels of stacking might be used, and the reliability consequences are not trivial. This work will discuss the results of several experiments designed to address the many aspects of successful PoP integration. PoP assembly can be performed as a single in-line process, where the bottom component is placed on a motherboard, and then the top component is placed onto the bottom, generally using a flux or paste dip process. However, there is also significant interest in “pre-stacking” the devices as an additional process, and then placing the complete module on the motherboard. Assembly processes such as flux dipping, paste dipping and pre-stacking were evaluated. PoP presents a unique challenge regarding warpage because of the multiple levels, so the warpage behavior was analyzed using Thermal Shadow-Moiré on each level as well as a pre-stacked module. The results correlate well with instances of head-in-pillow failures on select devices. Mechanical reinforcement through the use of underfill was also evaluated. Underfill was applied to the full module, as well as selectively to the top or bottom package. Corner bonding was also evaluated which provides mechanical reinforcement while still allowing the parts to be reworked. Both thermal cycle and mechanical reliability testing was undertaken to determine the effectiveness of the various parameters investigated. Finally, failure analysis was performed on select samples in order to determine the primary failure modes and locations, which were found to vary with the given test parameters.|
|Brian Roggeman, Process Research Engineer
Universal Instruments Corporation
Conklin, NY , NY