EMRS | alumni

The need to dissipate heat that results from high power applications using Flip-Chip Ball Grid Array (FC-BGA) packages has necessitated the attachment of heat spreaders. Knowing the deleterious effects of heat, electronic packages must be thermally enhanced to address the tradeoffs of efficiency, reliability, manufacturability, and cost versus thermal performance. This thermal solution can be accomplished by using various kinds of interface materials that have been evaluated as a part of this research. This research endeavor comprises a set of statistically designed experiments that were conducted and analyzed to derive meaningful results for the attachment of heat spreaders to FC-BGA packages. The integration of materials with disparate properties required rigorous testing and process refinements of heat spreader interface materials, rim seal adhesives, fluxes and encapsulants. The underlying philosophy in this approach is to achieve high reliability without compromising manufacturability. An initial set of experiments was targeted towards the identification of pertinent issues to be pursued. Several attach methods were explored and adhesion to silicon, copper, and solder mask was tested using Liquid-to-Liquid Thermal Shock (LLTS). Case studies addressed the effects of variations in the attachment of the heat spreader to the substrate on the warpage of the assembly during reflow. Optimization of the process for the attachment of a heat spreader to a flip chip package depends on the combination of the thermal interface and the rim seal used and, undoubtedly, on the package and heat spreader design itself. Therefore, the heat spreader attach process was developed for a product relevant test vehicle. A process with a reasonably broad ¡®window¡¯ was developed, ensuring that warpage induced variations in rim seal thickness did not appreciably affect the thermal interface thickness. The various attach methodologies foster different kinds of process damages that could be translated to the package. Irrespective of the approach, certain stresses, although temporarily, are applied to the package by virtue of attaching a heat spreader. Flip chip assemblies are inherently warped and twisted at room temperature and loads applied during the heat spreader attach process may partially or fully flatten them. Sensitivity to torsion and its impacts were therefore systematically studied. Loading and deformation during heat spreader attach might also conceivably damage the flip chip underfill. This work also investigated the potential repercussions of this damage to the underfill in reducing its resistance to subsequent moisture exposure and thermal excursions. Finally, designed experiments were conducted to evaluate the reliability of six different approaches of rim seal-thermal interface combinations. Responses such as shear strength (at time-zero and after thermal cycling), delamination at the thermal interface, package warpage, and heat spreader standoff measurements before and after cycling were measured and analyzed.


	sitemap \| contact us