Underflow Process for Direct-Chip-Attachment Packaging
- PDF / 1,509,810 Bytes
- 6 Pages / 417.6 x 639 pts Page_size
- 55 Downloads / 165 Views
Abstract--In flip-chip packaging an underfill mixture is placed into the chip-to-substrate standoff created by the array of solder bumps, using a capillary flow process. The flow behavior is a complex function of the mixture properties, the wetting properties, and the flow geometry. This paper reports on the use of a plane channel capillary flow to characterize underfill materials. The measured flow behavior provides evidence that both the contact angle (0) and the suspension viscosity (ppp) vary with time under the Influence of changing flow conditions. This nonlinear fluid behavior is modeled for the flow of both model suspensions and commercial underfill materials using an extended Washburn model. I. INTRODUCTION A major obstacle to the implementation of flip chip technology on a wide scale is the variability of the underfill process, in particular the lack of understanding of the basic mechanisms of underfill flow. The need for an understanding of the underfill process is magnified as many in surface mount technology consider switching to this flip chip technology. An understanding of the basic mechanisms of underfill flow, with predictive capability, would allow more rapid, widespread adoption of this efficient packaging technology. The major thrust of our research is to characterize the metrics which epitomize the basic mechanisms of underfill flow, in order to develop such a capability. For protection of both the chip and the solder joints in direct-chip-attachment (DCA), an epoxy resin underfills the standoff region between the chip and substrate (Fig.l). A capillary flow process is used to draw the underfill into this region. To achieve reliable fatigue life of the solder joint, the coefficient of thermal expansion of the cured underfill must approach that of the solder. A mixture with the desired, effective coefficient of thermal expansion is commonly obtained by adding a high volume fraction of solid particles to an epoxy. The effects of such fillers on the mechanisms of the capillary flow process are not well understood. Therefore we are conducting an investigation of the flow SO= • process that is used for underfill materials k~kW in DCA packaging. We seek to identify the means of achieving higher underflow speeds, and to develop the proper metrics for characterizing these flow processes. -A Experiments are performed to assess the effects on underfill flow of variations in "4Bo•d parameters describing the particles, the liquid carrier, and aspects of the flow Figure 1: A sketch of an underfilled package, geometry. These parameters, and related including the gap between chip and substrate, and system variables, are shown in Table One. the array of solder balls.
_______
117
Mat. Res. Soc. Symp. Proc. Vol. 515 01998 Materials Research Society
Particles
Carrier Fluid
System
Density pp Diameter a Volume Fraction
Density pf Viscosity g. Surface Tension a
Temperature T Plate Spacing s Contact Angle 0 Surface
Table One: The system state variables for the capillary flow of particle suspensions.
The
Data Loading...
