-
- News
- Books
Featured Books
- smt007 Magazine
Latest Issues
Current IssueIPC APEX EXPO 2024 Pre-show
This month’s issue devotes its pages to a comprehensive preview of the IPC APEX EXPO 2024 event. Whether your role is technical or business, if you're new-to-the-industry or seasoned veteran, you'll find value throughout this program.
Boost Your Sales
Every part of your business can be evaluated as a process, including your sales funnel. Optimizing your selling process requires a coordinated effort between marketing and sales. In this issue, industry experts in marketing and sales offer their best advice on how to boost your sales efforts.
The Cost of Rework
In this issue, we investigate rework's current state of the art. What are the root causes and how are they resolved? What is the financial impact of rework, and is it possible to eliminate it entirely without sacrificing your yields?
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - smt007 Magazine
New Solder Bumping Technology and Adapted Assembly Processes for 100 µm Pitch Flip-Chip Technology Using Capillary Flow or No-Flow Underfill, Part II
September 14, 2010 |Estimated reading time: 10 minutes
Editor's Note: To read Part I of this paper, click here.Results and Data Obtained Wafer Bumping Process
For the wafer bumping process, the yield is of highest importance. With WLSST technology, 60 µm solder spheres can be placed with a yield of 99.8% per Wafer. A uniform height and diameter of the bumps on the flip-chip are very relevant for the following process steps and especially have an influence on the reliability of the solder joints. Therefore the height and diameter of the solder bumps have been measured using a Werth Videocheck IP 400HA system. Exemplary the results for one side of a chip is shown for 60 µm and 50 µm solder spheres. As can be seen a sphere diameter of 60 µm results an average bump diameter of 66.5 µm (standard deviation: 0.8 µm) and an average height of 44.8 µm (standard deviation: 0.5 µm). Accordingly, a 50 µm solder sphere in initial state results a solder bump with an average diameter of
61.8 µm (standard deviation: 0.9 µm) and an average height of 36.3 µm (standard deviation: 0.7 µm). This shows that for both, the 60 µm and the 50 µm solder spheres very uniform solder bumps can be achieved on the flip-chip.
Layout and Substrate Structuring
To improve the yield during assembly and underfill of the flip-chips, several improvements on the structuring of the metallization and the solder mask of the test coupons have been made. As described before preliminary tests have shown that a solder mask is absolutely essential for high yields after reflow. The quality of the metallization is shown in Figure 6. As can be seen very precise structures could be realized on both the FR4 and BT substrates. Even ultra fine pitched structures with lines of 20 µm are of high accuracy.Figure 6: Quality of the ultra fine pitch structures (Layout V3.3 before solder mask application).
Regarding the solder mask quality the registration to the metallization, the height of the mask and the web width are of highest importance. All have direct influence on the long term reliability of the solder joints. A precise registration and web widths are needed for uniform footprints. A solder mask height as small as possible reduces the gap between the mask and the bottom side of the flip-chip what enhances the flow of the underfill.Figure 7: Accuracy of pad and solder mask.Two different foot prints and solder mask openings have been realized. On the one hand a tear drop design to support X-ray inspection (V3.2), on the other hand a conventional design to form solder joints as constant as possible (V3.3 and V3.3/1). Both designs show very good registration as can be seen in Figure 7.
The dimensions of the metallization and the solder mask are shown in Figure 8 in a cross section view for the 60 µm structures. The structuring of the metallization results in a line width of 46.7 µm (nominal 50 µm) and a height of 9.4 µm. Solder mask height is 25 µm.Figure 8: Cross section view of the metallization (60 µm structures) and the solder mask opening (V3.3).
Assembly Process
For flip-chip assembly the production yield is of high importance as well. Therefore the functionality of the assembled dies was tested after reflow and underfill, respectively Table 6. On the layout with the teardrop design (V3.2) a yield of 89.6 % for the flip-chips with 60 µm solder bumps and 81,3 % for the dies with 50 µm solder bumps could be achieved. Although the yield is quite good, a better yield is limited to the tolerances in the solder mask. For theses highly miniaturized structures it is essential that the registration of the solder mask fits precisely to the metallization. And even though the solder mask registration shows only tolerances of a few microns the yield suffers from the very fine structures of the solder mask in layout V3.2.
Table 6: Comparisons of the yield after reflow (capillary flow underfill process).If the openings in the solder mask are increased, the tolerance of the registration of the solder mask to the metallization becomes less critical. With layout V3.3, the yield could be raised to 93.1 %. With an improved printed board version (V3.3/1), a yield after reflow of 100 % has been achieved in fully automated assembly of the flip-chips.
As yet in the no flow underfill process a yield of 57% could be achieved with underfill A on the layout v3.3. As the research is still underway even better yields can be expected, especially with the improved layout V3.3/1.
Reliability Testing
The reliability testing of the test coupons of experiment 1 are carried out according to MIL-STD883G, Method 1010.7, Condition B and EIA/JESD22-A101-B. Figure 9 shows the Weibull plot for printed board version 3.2 with chips with 60 micron solder spheres at the first row. The capillary flow underfill B has been used for these experiments. As can be seen 51% of the dies show still no failure after 4,000 cycles. With this result, a characteristic life of 5,285 cycles can be calculated using the maximum likelihood method. The form factor of 1.16 indicates a process in steady state. But 24% of all failures on the tested coupons occur before 1,500 cycles and can be considered early failures. The reason for those early failures could be traced back to the very small openings of the solder mask of layout V3.2. The slightest variation from the nominal dimension of the opening leads to a solder joint of irregular shape. This again leads to an uneven strain distribution in the solder joint causing an early failure.
The reliability tests with printed boards from experiment 2 and 3 are still in progress. According to the present knowledge from the results of the tests a similar reliability of the solder joints can be expected.
The reliability testing of the test coupons from experiment 4 and 5 (underfill A) was carried out according to DIN EN 60 068-2-14. One thousand cycles were performed. Figure 11 shows the Weibull plot for those experiments. The calculated characteristic life of the components is 2,067 cycles, what is still considered a good result for those extremely high miniaturized assemblies. As can be seen, over 70% of the failures of the test coupons with 60 µm solder joints happen before 300 cycles. The assumption for the root cause of the early failures is supported by the form factor, which has a value of 0.8. The reason for that is described above and can again be traced back to the solder mask. This becomes obvious when looking at the results for the 50 µm solder spheres. For those even smaller structures in the solder mask even the slightest tolerances become critical in terms of the reliability of the solder joints. This leads to a relatively low characteristic life of 977 cycles. The form factor of 1.97 indicates that the defects are no early failures. But again, with the improved layout V3.3/1 notable better results are to be expected during life cycle testing.Figure 9: Weibull plot for printed board version 3.2 with chips with 60 micron solder spheres at the first row (experiment no. 1).Figure 10: Cross-section view of assembled flip-chip with 60 µm solder spheres.Figure 11: Weibull plot for the comparison between 60 µm and 50 µm solder spheres (experiments 4 and 5).A characteristic sign for a failure of the flip-chip is shown in Figure 12. During hold time at high and low temperature, respectively, the voltage increases and decreases slightly as can be clearly seen in the left side of the upper diagram in Figure 12. Through this significant signal difference, the number of temperature cycles can be identified. The drop of the signal during temperature change indicates the beginning of a crack in the solder joint. After several drops the signal starts jittering at high temperatures, indicating crack growth. During the course of temperature cycling, the signal starts jittering at low temperatures as well, leading finally to a failure of the solder joint.Figure 12: Signal from online measuring during temperature cycling.
Figures 10 and 13 show exemplary cross-section views of the solder joints of test coupons with capillary flow underfill. The solder joint on the left of Figure 13 shows a slight bend defined by the solder mask. Such an asymmetry can lead to accelerated cracking of the interconnection.Figure 13: Cross-section view of the solder joints.ConclusionIn conclusion, we have presented a new low-cost manufacturing technology for ultra fine pitch flip-chip applications based upon standard processes of the surface mount technology. Solder spheres with a diameter of 60 µm, as well as 50 µm, can be placed onto wafers simultaneously using the highly efficient and flexible wafer level solder sphere transfer process also known as gang ball placement with very high yield. The automatic assembly of the flip-chips has been demonstrated. The yield of the flip-chip process is depending from layout and tolerances of the printed boards, especially from solder mask registration and solder mask tolerances. With the final printed board layout we obtained 100% assembly yield (n=32) after underfill. The root cause for early failures during temperature cycling could be identified and will be conducive to improved reliability results using printed boards with the final layout. Underfill selection and underfill process optimization are crucial for the long term reliability of those extremely high miniaturized assemblies.
Acknowledgements The authors would like to thank Thomas Friedrich from MSE for automatic assembly of the chips and Bernd Burger from MSE for cross sectioning several test coupons. The results presented in this paper have been achieved within the project Process Technology and Connection Methods for Ultra Fine Pitch Components. The authors would like to thank the Federal Ministry of Education and Research of Germany for funding under contract 02PG2361, 02PG2362, 02PG2363 and 02PG2366.
References:1. A. Strandjord et al., "WLCSP Production Using Electroless Ni/Au Plating and Wafer Level Solder Sphere Transfer Technology," International Wafer-Level Packaging Conference 2008, San Jose, California.2. D. Manessis et al., "Technological advancements in Lead-free Wafer Bumping using Stencil Printing Technology," European Microeletronics and Packaging Conference, Brugge, Belgium, pp. 427-433, 2005.3. D. Manessis et al., "Evaluation of Innovative Nano-Coated Stencils in Ultra-Fine-Pitch Flip Chip Bumping Processes," IMAPS 2008 41st International Symposium on Microelectronics, Providence, Rhode Island.4. M. Klein et al., "Development and Evaluation of Lead Free Reflow Soldering Techniques for the Flip Chip Bonding of Large GaAs Pixel Detectors on Si Readout Chip," 58th Electronic Components and Technology Conference 2008, Lake Buena Vista, Florida, pp. 1893-1899.5. T. Oppert et al., "Laser assisted soldering and Flip-Chip attach for 3-D packaging," 31st International Conference on Electronics Manufacturing and Technology 2006, Petaling Jaya, Malaysia.6. T. Teutsch, "Solder Ball Placement as an Alternative to Stencil Printing for Flip Chips and Wafer-Level Packaging," Chip Scale Review, August-September, 2005.7. B. Dang et al., "50µm Pitch Pb-Free Micro-bumps by C4NP Technology," 58th Electronic Components and Technology Conference 2008, Lake Buena Vista, Florida, pp. 1505-1510.8. E. Laine et al., "C4NP as a High-Volume Manufacturing Method for Fine-Pitch and Lead-Free Flip Chip Solder Bumping," 1st Electronics Systemintegration Technology Conference, Dresden, Germany, pp. 518-524.9. M. Fendler et al., "A Fluxless and Low-Temperature Flip Chip Process Based on Insertion Technique," IEEE Transactions on Components and PackagingTechnologies, Vol. 32, No 1; March, 2009, pp. 207-215.10. A. Tanaka et al., "Study of stress to Solder Joint by Underfill filling," 41st International Symposium on Microelectronics 2008, Providence, Rhode Island.11. G. Carson, M. Edwards, "Factors Affecting Voiding in Underfilled Flip Chip Assemblies," LOCTITE Technical Paper, February, 2002.12. P. N. Houston, B. A. Smith, D. F. Baldwin, "Implementation of No Flo Underfill Materials for Low Cost Flip Chip Assembly," SMTA International, 2001.13. R. Thorpe, D. F. Baldwin, "Yield Analysis and process Modeling of Low Cost, High Throughput Flip Chip Assembly Based on No Flow Underfill Materials," IEEE Transaction and Electronics Packaging Manufacturing, Vol. 24, No. 2, April, 2001, pp.123-135.14. C. P. Wong, S. H. Shi, G. Jefferson, "High Performance No-Flow Underfills for Low-Cost Flip-Chip Applications: Material Characterization," IEEE Transactions on Component, Packaging and Manufacturing Technology-Part A, Vol. 21, No. 3, September, 1998, pp.4501-458.15. C.-M. Liu, K.-N. Chiang, "Solder Shape Design and Thermal Stress/Strain Analysis of Flip Chip Packaging using Hybrid Method," International Symposium on Electronic Material & Packaging, 2000, Hong Kong, pp. 44-50.16. K.-N. Chiang, Y.T. Lin, H.-C. Cheng, "On Enhancing Eutectic Solder Joint Reliability Using a Second-Reflow-Process Approach," IEEE Transactions on Advanced Packaging, Vol. 23, No. 1, February, 2000, pp.9-14.17. J. H. Lau, C. Chang, S.-W. R. Lee, "Failure Analysis of Solder Bumped Flip Chip on Low-Cost Substrates," IEEE Transactions on Electronics Packaging Manufacturing, Vol. 23, No. 1, January, 2000.18. C.-T. Peng et al., "Reliability Analysis and Design for the Fine-Pitch Flip Chip BGA Packaging," IEEE Transactions on Components and Packaging Technologies, Vol. 27, No. 4, December, 2004, pp. 684- 692.19. A. Strandjord et al., "WLCSP Mechanical Reliability - High Speed Pull Testing (Lead-Free Solder Alloys and Electroless Nickel UBM)," IMAPS 2008 41st International Symposium on Microelectronics, Providence, Rhode Island, pp. 108-114.