打線接合之線材漸漸由銅線取代金線，但是製程設定也將會受到銅線的機械性質所影響與限制。銅線的硬度相對於金線高出許多，在打線過程中需要調整製程參數的設定，才能使接合品質達到需求標準。
首先，本文使用ANSYS12/LS-DYNA對打線接合進行三維非線性動態模擬，分析打線接合過程中之動態與力學行為，並確認模型之可靠度。隨後，探討製程參數因子，包含：接合力量、超音波振幅、超音波頻率與接合速度等參數，對於打線結構之常見缺陷生成的原因及影響力，如銅銲球頸縮、鋁墊殘餘厚度過小、鋁擠出寬度所造成的銲墊結構失效與銲墊內部low-K層的損壞。因此，經分析發現當接觸速度為27.5mm/s時可有效的降低衝擊瞬間於low-K層所產生的應力，及接合力量與超音波振幅為顯著因子。接著繼續探討使用兩階段力量負載之效應與優點。發現使用兩階段力量負載可以有效的降低超音波階段於low-K層外側點所產生的應力與銅銲球頸部緊縮程度，尤其在大超波振幅之下，分別由202.24MPa降至160.97MPa與1.8997μm降至1.1749μm，甚至找到一個第二階段力量負載之最佳值(31.5gf)，可減少超音波階段鋁向外推擠之程度。最後，歸納出一套調配製程參數之準則，以達到提升銅銲線製程之可靠度。其結果顯示降低超音波振幅、第一階段力量負載與降低第二階段力量負載，可以改善銅銲球頸部緊縮所導致的斷線行為。超音波振幅由2μm調降至0.25μm、銅銲球頸部緊縮於超音波階段由1.8997μm降至0.0282μm。第一階段力量負載由45.5gf降至24.5gf，銅銲球頸部緊縮於接合瞬間縮由3.6299μm降至2.1754μm。第二階段力量負載由35gf調降至21gf，銅銲球頸部緊縮於超音波階段由0.835μm降至0.7482μm；降低超音波振幅與降低第一階段力量負載，可以改善鋁墊殘餘厚度過小所導致電性可靠度下降之問題。超音波振幅由2μm調降至0.25μm，鋁墊最低點下壓深度由0.8495μm改善至0.6553μm。第一階段力量負載由45.5gf調降至24.5gf，鋁墊最低點下壓深度由0.8598μm改善至0.5023μm；鋁墊向外推擠造成SiN PSV的破壞，可藉由降低超音波振幅及降低第一階段與尋找最佳的第二階段力量負載，以獲得改善。超音波振幅由2μm調降至0.25μm，鋁墊的向外推擠由56.689μm減至51.804μm。當第二階段力量負載為28gf時，有一最佳值53.389μm；若衝擊瞬間之力量過大導致low-K層中心點之損壞，可藉由降低第一階段力量負載及尋找最佳的接觸速度，以獲得改善。第一階段力量負載由45.5gf降至24.5gf，中心點最大應力由233.822MPa降至140.201MPa。當接觸速度為22.5mm/s時，有一最佳值176.47MPa。然而外側點之破壞通常發生於超音波階段，可藉由降低超音波振幅及降低第二階段力量負載，以獲得改善。超音波振幅由2μm調降至0.25μm，外側點最大應力由202.241MPa降至126.354MPa。第二階段力量負載由35gf調降至21gf，外側點最大應力由159.774MPa降至144.243MPa。

關鍵字：銅打線接合、三維非線性動態模擬、頸縮、鋁擠出、應力分析

In recent years, the Cu wire has gradually replaced the Au wire and becomes the popular material for the wire bonding. Thus the process parameters of wire bonding should be affected by the material properties of the Cu wire, i.e., the hardness of the Cu wire is significantly higher than that of the Au wire. In order to achieve the bonding quality, the adjustment and setting of the process parameters in the bonding process are required.
First of all, the Finite Element Analysis software ANSYS12.0/LS-DYNA is adopted for 3D transient nonlinear dynamic simulation. The dynamic and mechanical behavior is analyzed and the reliability of the model is verified. Next, the effects of process parameters including bonding force, ultrasonic amplitude, frequency, and contact velocity on the formation of defects are investigated. For instance, the structural inefficiency of the bonding pad resulting from neck reduction of Cu ball, residual of Al pad and Al push-out, and the stress on the low-K layer are investigated. It is found that the stress on the low-K layer at the bonding impact moment can be effectively reduced with the contact velocity of 27.5mm/s. Besides, the ultrasonic amplitude and the bonding force are recognized as significant parameters.
Secondly, effects and advantages of the double load technology are applied and analyzed. The results show that the application of the double load technology can effectively reduce the stress on the peripheral low-K area and the neck reduction of Cu ball. In particular, at the large ultrasonic amplitude, the stress on the peripheral low-K area as well as the neck reduction of Cu ball are found to reduce from 202.24MPa to 160.97MPa and 1.8997μm to 1.1749μm, respectively. Furthermore, an optimal second stage bonding force of 31.5gf is found so that the width of Al push-out can be reduced.
Finally, a set of guidelines for tuning the process parameters to achieve the reliability of the Cu wire bonding process is proposed. The results show that lower ultrasonic amplitude and impact force, and second stage bonding force can improve the behavior of bonding break due to the neck reduction of Cu ball. In other words, with the reduces of the ultrasonic amplitude from 2μm to 0.25μm, the impact force from 45.5gf to 24.5gf and the second stage bonding force from 35gf to 21gf, the necking reduces from 1.8997μm to 0.0282μm in the ultrasonic stage, from 3.6299μm to 2.1754μm at bonding impact moment, and from 0.835μm to 0.7482μm in the ultrasonic stage, respectively. Furthermore, reduce of ultrasonic amplitude and impact force can improve the problem of the weaker reliability of electricity caused by over thinness of Al pads. In addition, by reducing the ultrasonic amplitude from 2μm to 0.25μm and the impact force from 45.5gf to 24.5gf, the depth at the lowest point of Al pads can be decreased from 0.8495μm to 0.6553μm and from 0.8598μm to 0.5023μm, respectively. Also, the damage of SiN PSV wall caused by aluminum push-out can be improved by reducing the ultrasonic amplitude and the impact force, then looking for an optimal second stage bonding force. Therefore, the ultrasonic amplitude reduces from 2μm to 0.25μm, the width of Al push-out reduces from 56.689μm to 51.804μm. Also, an optimal value 53.389μm is found with the second stage bonding force of 28gf. In addition, reducing the impact force and looking for an optimal contact velocity can improve the damage on the central low-K area caused by larger impact force. For example, reducing the impact force from 45.5gf to 24.5gf can decrease the stress from 233.822MPa to 140.201MPa. Meanwhile, an optimal value 176.47MPa is obtained at the contact velocity of 22.5mm/s. However, the damage on the peripheral low-K area usually occurring at the ultrasonic stage can be improved by reducing the ultrasonic amplitude and second stage bonding force. Hence, by reducing the ultrasonic amplitude from 2μm to 0.25μm and the second stage bonding force from 35gf to 21gf, the stress can be decreased from 202.241MPa to 126.354MPa and from 159.774MPa to 144.243MPa, respectively.

Keywords: Copper wire bonding, 3D transient nonlinear dynamic simulation, Necking, Aluminum push-out, Stress Analysis

[1] Vern H. Winchell, "An Evaluation of Silicon Damage Resulting from Ultrasonic Wire Bonding," in Reliability Physics Symposium, 1976. 14th Annual, 1976, pp. 98-107.
[2] Jin Onuki, Masahiro Koizumi, and Isao Araki, "Investigation of the Reliability of Copper Ball Bonds to Aluminum Electrodes," Components, Hybrids, and Manufacturing Technology, IEEE Transactions on, vol. 10, pp. 550-555, 1987.
[3] Satoru Mori, Hideaki Yoshida, and Naoki Uchiyama, "The development of new copper ball bonding-wire," in Electronics Components Conference, 1988., Proceedings of the 38th, 1988, pp. 539-545.
[4] Kenji Toyozawa, Kazuya Fujita, Syozo Minamide, and Takamichi Maeda, "Development of copper wire bonding application technology," Components, Hybrids, and Manufacturing Technology, IEEE Transactions on, vol. 13, pp. 667-672, 1990.
[5] 陳家旭, "打線接合之實驗與有限元素研究," 碩士, 機械工程系, 國立交通大學, 新竹市, 2002.
[6] Hyoung-Joon Kim, Joo Yeon Lee, Kyung-Wook Paik, Kwang-Won Koh, Jimhee Won, Sihyun Choe, et al., "Effects of Cu/Al intermetallic compound (IMC) on copper wire and aluminum pad bondability," Components and Packaging Technologies, IEEE Transactions on, vol. 26, pp. 367-374, 2003.
[7] N. Srikanth, S. Murali, Y.M. Wong, and Charles J. Vath III, "Critical study of thermosonic copper ball bonding," Thin Solid Films, vol. 462, pp. 339-345, 2004.
[8] F. W. Wulff, C. D. Breach, D. Stephan, Saraswati, and K. J. Dittmer, "Characterisation of intermetallic growth in copper and gold ball bonds on aluminium metallization," in Electronics Packaging Technology Conference, 2004. EPTC 2004. Proceedings of 6th, 2004, pp. 348-353.
[9] Yoon-joo Kim, Joon-soo Kim, Ji-young Chung, Seok-ho Na, Jin-young Kim, and Seok-bong Kim, "Low-K wire bonding," in Electronic Components and Technology Conference, 2006. Proceedings. 56th, 2006, p. 7 pp.
[10] Chang-Lin Yeh, Yi-Shao Lai, and Chin-Li Kao, "Transient simulation of wire pull test on Cu/low-K wafers," Advanced Packaging, IEEE Transactions on, vol. 29, pp. 631-638, 2006.
[11] Vincent Fiori, Lau Teck Beng, Susan Downey, Sebastien Gallois-Garreignot, and Stephane Orain, "3D Multi Scale Modeling of Wire Bonding Induced Peeling in Cu/Low-k Interconnects: Application of an Energy Based Criteria and Correlations with Experiments," in Electronic Components and Technology Conference, 2007. ECTC '07. Proceedings. 57th, 2007, pp. 256-263.
[12] Yong Liu, Scott Irving, and Timwah Luk, "Thermosonic Wire Bonding Process Simulation and Bond Pad Over Active Stress Analysis," Electronics Packaging Manufacturing, IEEE Transactions on, vol. 31, pp. 61-71, 2008.
[13] C.J. Hang, C.Q. Wang, M. Mayer, Y.H. Tian, Y. Zhou, and H.H. Wang, "Growth behavior of Cu/Al intermetallic compounds and cracks in copper ball bonds during isothermal aging," Microelectronics reliability, vol. 48, pp. 416-424, 2008.
[14] Weidong Huang, "Computational modeling and optimization for wire bonding process on Cu/low-K wafers," in Electronic Packaging Technology & High Density Packaging, 2009. ICEPT-HDP '09. International Conference on, 2009, pp. 344-352.
[15] I. Qin, A. Shah, C. Huynh, M. Meyer, M. Mayer, and Y. Zhou, "Effect of process parameters on pad damage during Au and Cu ball bonding processes," in Electronics Packaging Technology Conference, 2009. EPTC '09. 11th, 2009, pp. 573-578.
[16] Xiangquan Fan, Kaiyou Qian, Techun Wang, Yuqi Cong, Mike Zhao, Binhai Zhang, et al., "Nanoindentation investigation of copper bonding wire and ball," in Electronic Packaging Technology & High Density Packaging, 2009. ICEPT-HDP '09. International Conference on, 2009, pp. 790-794.
[17] Meijiang Song, G.L. Gong, J. Z. Yao, Sean Xu, Stephen Lee, M.C. Han, et al., "Study of optimum bond pad metallization thickness for copper wire bond process," in Electronics Packaging Technology Conference (EPTC), 2010 12th, 2010, pp. 597-602.
[18] Takashi Hisada, Toyohiro Aoki, Keishi Okamoto, Shinichi Harada, John C. Malinowski, Keith F. Beckham, et al., "Fine Pitch Wirebonds on Ultra Low-k Device," in Transactions of The Japan Institute of Electronics Packaging 4(1), 17-23, 2011
[19] Leong Ching Wai, N.B. Jaafar, M. Chew, Sivakumar, Gunasekaran, Kanchet, et al., "Fine pitch copper wire bonding on 45nm tech Cu/low-k chip with different bond pad metallurgy," in Electronics Packaging Technology Conference (EPTC), 2011 IEEE 13th, 2011, pp. 752-757.
[20] Dominik Stephan, Effie Chew, Johnny Yeung, and Eugen Milke, "Impact of palladium to the interfacial behavior of palladium coated copper wire on aluminium pad metallization during high temperature storage," in Electronics Packaging Technology Conference (EPTC), 2011 IEEE 13th, 2011, pp. 294-303.
[21] Zhaohui Chen, Yong Liu, and Sheng Liu, "Modeling of copper wire bonding process on high power LEDs," Microelectronics Reliability, vol. 51, pp. 171-178, 2011.
[22] John D. Beleran, Yang Yong Bo, Hyman G. Robles, Antonino Milanes, Alfred Yeo, and Chan Kai Chong, "Copper wire bond analysis: Pad design effects and process considerations," in Electronic Components and Technology Conference (ECTC), 2012 IEEE 62nd, 2012, pp. 1124-1129.
[23] Y.B. Yang, Kumar N., John D., Hyman R., Clive F., Nathapong S., et al., "A dynamic study of pad structure impact on bond pad/low-K layer stress in copper wire bond," in Electronics Packaging Technology Conference (EPTC), 2012 IEEE 14th, 2012, pp. 654-658.
[24] 林冠中, "以ALE與外顯示時間處理法進行LED打線接合製程之模擬分析," 碩士, 動力機械工程學系, 國立清華大學, 新竹市, 2013.
[25] 余健源, "銅線於銲線接合製程受衝擊之最適設計," 成功大學工程科學系學位論文, pp. 1-162, 2014.
[26] John O. Hallquist, LS-DYNA Theory Manual, Livermore Software Technology Corporation, 7374 Las Positas Road Livermore California 94551, March 2006.
[27] A. OBrien and C. Guzman, ed. 2007, Welding Handbook, 9th Edition, Volume 3
[28] 張巍耀，“金線與銅線在熱影響區的材料特性及其應用於銲線製程之動態分析”，義守大學機械與自動化工程學系碩士論文，高雄，台灣，2008
[29] ANSYS Coupled-Field Analysis Guide