Semiconductor device manufacturing: process – Packaging or treatment of packaged semiconductor – Assembly of plural semiconductive substrates each possessing...
Reexamination Certificate
1999-11-10
2002-11-05
Talbott, David L. (Department: 2827)
Semiconductor device manufacturing: process
Packaging or treatment of packaged semiconductor
Assembly of plural semiconductive substrates each possessing...
C438S613000, C228S180220
Reexamination Certificate
active
06475828
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to packages for integrated circuits and more particularly to a method for dispensing underfill during the manufacture of a flip-chip package.
2. Description of the Related Art
A flip-chip package comprises an integrated circuit die whose active surface (i.e. the surface with bond pads) faces bond pads on a substrate. The bond pads on the integrated circuit are connected with corresponding bond pads on the substrate through solder bumps, which may be formed on the bond pads of the die. Underfill is dispensed between the solder bumps and the substrate to improve the solder fatigue life of the solder bumps.
According to a first conventional method, the die is first “flipped” on to the substrate such that the solder bumps contact the corresponding bond pads on the substrate. The solder balls may then be reflowed. Underfill epoxy may then be dispensed between the die and the substrate to fill the gaps between the solder balls. A typical method involves dispensing the underfill from one side of the substrate to fill the other side; capillary action pulls the material across the substrate through the gaps between the solder bumps.
After reflow, the bump height is less than the original bump height. In a typical flip chip, the bump height may decrease by 25 um, which in turn decreases the gap height between a solder mask (which is overlaid on the substrate) and the die. This reduction in the gap between a chip and a substrate increases the difficulty of dispensing the underfill; specifically, a smaller gap impedes the flow of underfill between the chip and the substrate and thereby increases the amount of voiding in the underfill material. Voiding in the underfill results in decreased reliability of the attachment of die to substrate under conditions of stress, such as temperature cycling and moisture preconditioning. Voids also allow for potential electrical shorts between adjacent bumps.
Solder masks have other drawbacks. For example, dimensional tolerances of the solder mask openings for the pads on the substrate can limit the density (pitch) between solder balls due to substrate manufacturing defects and defects involved in the assembly of the flip chip. These defects lower manufacturing yields and increase the cost of the substrate and the cost of flip chip assembly.
Alternatively, according to a second conventional method, a “no flow flux-underfill” may be used as an underfill. In this case, an underfill with a fluxing component is dispensed on the substrate and the die is then “flipped” onto the substrate as previously described; since the underfill is dispensed before the attachment of the die to the substrate, the second conventional method avoids the problems of dispensing underfill between the gap of a die already attached to a substrate. The resulting assembly is then run through a reflow furnace to reflow the solder balls and cure the flux-underfill material. The major drawback to the flux-underfill method is that when the flux is loaded with filler, the solder does not properly reflow. On the other hand, when no fillers are used, the flux has high CTE and low Tg, which results in poor reliability, especially in temperature cycling tests.
As the pitch of solder bumps continues to decrease (i.e. the density of bond pads on dies increases), and therefore the size of solder bumps decreases, gaps between substrates and dies will continue to decrease. Therefore, it will become increasingly important to overcome the aforementioned limitations in the above mentioned underfill dispense techniques.
SUMMARY OF THE INVENTION
The present invention comprises a method for applying underfill to a flip-chip package comprising an integrated circuit die and a substrate. The method comprises the following steps. First, a first non-filled no flow flux-underfill is applied to a plurality of solder bumps disposed on an active surface of the integrated circuit die. Next, the integrated circuit die is placed on the substrate such that the solder bumps align with corresponding bond pads on the substrate, thereby creating an assembly. A second filler-loaded no flow flux-underfill is then dispensed on a side of the substrate such that the second filler-loaded no flow flux-underfill flows by capillary action between the die and substrate to fill a gap therebetween. Finally, the assembly is passed through a furnace such that both the first non-filled no flow flux-underfill and the second filled no flow flux-underfill are cured, and such that the solder bumps are reflowed.
The above described method has the advantages of using a flux as an underfill without the previously described disadvantages associated with the second conventional method.
REFERENCES:
patent: 5376588 (1994-12-01), Pendse
patent: 5385869 (1995-01-01), Liu et al.
patent: 5696027 (1997-12-01), Crane, Jr.
patent: 5880530 (1999-03-01), Mashimoto et al.
patent: 5952726 (1999-09-01), Liang
patent: 5985043 (1999-11-01), Zhou et al.
patent: 5988485 (1999-11-01), Master et al.
patent: 5998242 (1999-12-01), Kirkpatrick et al.
patent: 6048753 (2000-04-01), Farnworth
patent: 6064113 (2000-05-01), Kirkman
patent: 6103549 (2000-08-01), Master et al.
patent: 6111756 (2000-08-01), Moresco
patent: 6117759 (2000-09-01), Greer et al.
patent: 6121689 (2000-09-01), Capote et al.
patent: 6198635 (2001-03-01), Shenoy et al.
patent: 6234379 (2001-05-01), Donges
patent: WO 01/88959 (2001-11-01), None
patent: WO 02/24391 (2002-03-01), None
Kulojarvi et al., “High Volume Capable Direct Chip Attachment Methods”, 1999 Electronic Components and Technology Conference, pp. 441-445.*
Smith et al., “A reliability and Failure Mode Analysis of No Flow Underfill Materials for Low Cost Flip Chip”, 2000 Electronic Components and Technology Conference, pp. 1719-1730.*
DeBarros et al., “The No-Flow Fluxing Underfill Adhesive For Low Cost, High Reliability Flip Chip Assembly”, 1999 Electronic Components and Technology Conference, pp. 955-960.*
wong et al., “Hish Performance No Flow Underfills for Low-Cost Flip-Chip Applications”, 1997 Electronic Components and Technology conference, pp. 850-858.*
Shi et al., “Development of the Wafer Level compressive-Flow Underfill Encapsulant”, 1999 International Symposium on Advanced Packaging Materials, pp. 337-343.*
Shi et al., “Recent Advances in the Development of No-Flow Underfill Encapsulants—A Practical approach towards the Actual Manufacturing Application”, 1999 Electronic Components and Technology Conference, pp. 770-776.*
Shi et al., “Development of the Wafer Level compressive-flow Underfill Process and Its Involved Materials”, 1999 Transactions of Electronics Packaging Manufacturing, vol. 22, No. 4, Oct. 1999, pp. 274-281.
Chambliss Alonzo
LSI Logic Corporation
Luedeka Neely & Graham PC
Talbott David L.
LandOfFree
Method of using both a non-filled flux underfill and a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of using both a non-filled flux underfill and a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of using both a non-filled flux underfill and a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2931182