對於現代積體電路封裝製程中的球銲過程而言，緊密的金-鋁接觸是產生良好介金屬化合物的重要條件。 這種良好的介金屬化合物能長時間地在內部的矽晶片及外部接點之間提供一個堅固而導電的通路。 然而，根據封裝業的數據收集的結果來看，並非所有的球銲介金屬化合物都足夠良好。 事實上，由於介金屬化合物不良而造成的球脫落，仍舊導致相當的良率損失及顧客抱怨。
在諸多造成球脫落的原因中，有兩個因素 – (1) 在晶圓製程中鋁墊回火的條件(操作區間)不良, 及(2) 在封裝製程中鋁墊表面遭受環境中的溼氣過高 而過度氧化所造成的影響仍舊不明確。 在本研究中，設計了數個實驗以評估上述兩因子對球銲品質的影響。 實驗的結果証實，在一般的情形下，此兩因子並不會造成球銲品質變差。
本研究的最後一部份為找出最佳的清潔方法以去除銲墊上的頑固污染物。 有時封裝廠的生產線上會發現某些不易去除的污染物，它們常常導致整批的晶圓都必需被報廢；意即損失大筆的金錢。 在實驗完成後，離子電漿刻蝕被証實為最適當的方法，而最佳刻蝕條件亦同時被確認。 最後，利用此法清潔完成的樣品也利用推球測試而被証實具有良好的可銲性。

Intimate contact between Au and Al is very important for forming a good Au-Al intermetallic during thermosonic Au-Al ball bonding in modern IC packaging process. This good intermetallic can provide strong and electrical connection between silicon chip, gold wire and external electrical terminal to keep IC device active for a long time. However, based on the data collection in IC packaging house, not all of intermetallic after ball bonding is good enough. Lifted ball bond due to poor intermetallic still causes big yield loss and a lot of customer complaints.
Among the possible causes of causing lifted ball bond, the influence of two factors – (1) Thicker oxidation resulted from improper metallzation annealing window in wafer process and (2) Bondpad over-oxidation due to humidity from manufactured environment are still unknown. Experiments were designed to evaluate their contribution in ball boning quality, and conclusion was summarized based on the test results – both factors would not weaken the ball bond strength under normal condition.
The last part of this study is to find the clean method and optimal condition for removing tough contamination on bondpad surface. This kind of contamination was observed sometimes in assembly center and most of them cause a lot of wafer to be scrapped, i.e. a lot of money is lost. Plasma etch is identified as the most suitable method, and the optimal condition was found after experiments. The bondability of cleaned bondpad was also confirmed by later ball shear test.

[1-1] George G. Harman, “Wire Bonding in Microelectronics”, second edition, McGraw-Hill, 1997, pp. 29-30.
[1-2] See, for example, Hansen, M., ”The Constitution of Binary Phase Diagrams, 2d Ed, McGraw-Hill, 1958, New York.
[1-3] Philofsky, E., “Intermetallic Formation in Gold-Aluminum System”, Solid State Electronics, Vol. 13, 1970, pp. 1391-1399.
[1-4] Majni, G., and Ottaviani, G., “AuAl Compound Formation by Thin Film Interactions”, J. Crystal Growth, Vol. 47, 1979, pp. 583-588.
[2-1] White, M. L., Serpiello, J.W., Stringy, K.M., and Rosenzweig, W., “The Use of Silicone RTV Robber for Alpha Particle Protection on Silicon Integrated Circuits”, 19th Annual Proc., reliability Physics, Orlando, Florida, April 7-9, 1981, pp. 43-47.
[2-2] George G. Harman, “Wire Bonding in Microelectronics”, second edition, McGraw-Hill, 1997, pp. 123-125.
[2-3] Gerling, W., “Electrical and Physical Characterization of Gold-Ball Bonds on Aluminum Layers”, IEEE ECC, New Orleans, Louisana, May 14-16, 1984, pp.13-20.
[2-4] George G. Harman, “Wire Bonding in Microelectronics”, second edition, McGraw-Hill, 1997, pp. 149-152.
[2-5] Harman, G. G., and Wilson, C.L., “Materials Problems affecting Reliability and Yield of Wire Bonding in VLSI Devices,” Proc. 1989 MRS, Electronic Packaging Materials Scince IV, Vol. 154, San Diego, California, March 30-31, 1982, pp.16-26.
[2-6] Clarke, R.A., and Lukatela, V., “Inadequacy of Current Mil-STD Wire Bond Certification Procedure,” Proc. 1991 Intl. J. Microcircuit andElectronic Packaging, Vol. 15, pp. 87-96
[6-1] Weiner, J. A., Clatterbaugh, G. V., Charles, H. K. Jr. and Romenesko, B. M., "Gold Ball Bond Shear Strength - Effects of Cleaning, Metallization, and Bonding Parameters," Proc. of the 33rd Electronics Components Conf., Oriando, Florida, May 16-18,1983, pp. 208-220.
[6-2] Sowell, R. R., Cuthrell, R. E., Mattox, D. M., and Bland, R. D., "Surface Cleaning by Ultraviolet Radiation," J. Vac. Sci. Technol., Vol. II, Jan/Feb., 1974, pp. 474-475.
[6-3] Mittal, K. L. (ed.). Surface Contaminator, Genesis, Detection, and Control, Plenum Press, New York, 1979, Vols. 1 and 2.
[6-4] Jellison, J. L., "Effect of Surface Contamination on the Thermocompression Bondability of Gold," IEEE Trans. on Parts, Hybrids, and Packaging, Vol. II, Sept. 1975, pp. 206-211.
[6-5] Holloway, P. H., and Bushmire, D. W., "Detection by Auger Electron Spectroscopy and Removal by Ozonization of Photoresist Residues," Proc. of the 12th Annual International Conf. on Reliability Physics, Las Vegas, Nevada, April 1974, pp.180-186.
[6-6] Vig, J. R., and Le Bus, J. W., "UV/Ozone Cleaning of Surfaces," IEEE Thins, on Parts, Hybrids, and Packaging, Vol. 12, Dec. 1976, pp. 36S-370.
[6-7] Jellison, J. L., and Wagner, J. A., "Role of Surface Contaminants in the Deformation Welding of Gold to Thick and Thin Films," 29th Electronics Components Conf., 1979, pp. 336-345.
[6-8] Clarke, F. K., "UV/Ozone Processing: Its Applications in the Hybrid Circuit Industry," Hybrid Circuit Technology, Dec. 1985, p. 42.
[6-9] Zafonte, L., and Chin, R., "UV/Ozone Cleaning for Organics Removal on Silicon Wafers," SPIE 1984 Microlithography Conf., Santa Clara, California, March 11-16, 1984, Paper No. 470-19.