覆晶載板接合封裝具有細間距、訊號傳遞快速、散熱性良好等優點，本文針對覆晶載板接合封裝進行研究，希望透過分析能提高封裝之疲勞壽命以確保產品之品質穩定。
本文先使用ANSYS有限元素分析軟體對覆晶載板接合封裝進行分析，模型施加低溫-25°C至高溫125°C的循環負載，其中錫球需考慮其為彈塑性材料，其他材料則視為彈性材料。為了提升模擬效率採用全域/局部方法，並以錫球累積之平均應變能密度，作為評估封裝整體可靠度之指標，然後藉由Syed 提出之疲勞壽命公式，預測覆晶載板接合封裝之疲勞壽命。其次，利用單一因子分析探討各控制因子對平均應變能密度值之影響，並以部分因子設計法篩選顯著之因子效應，再將其設計參數建立迴歸模型，然後引用基因演算法進行最佳化分析。
最後，利用區間式基因演算法，可得知各參數對可靠度指標的敏感度，其順序由大至小分別為底填膠熱膨脹係數、銅柱直徑、電路板厚度、銅柱高度。但實際上還會受到製程規格的限制，因此必須界定成本與技術允許下製程參數的區間範圍，使其區間內各種參數組合所得之反應值皆在允許誤差範圍內，以確保產品的品質穩定。

The Flip Chip on Board is recognized to have the advantages of fine pitch, rapid signal transmission and well heat dissipation. By adopting the Flip Chip on Board, this paper aims to improve the fatigue life of the package and ensure the stability of product quality.
First of all, the ANSYS finite element analysis software is applied. The model of package is given with the thermal cycle of -25°C ~ 125°C, in which the solder ball is considered as elastoplastic while other components are elastic. The global/local method is applied to promote the simulated efficiency. The average strain energy density accumulated in the solder ball is treated as the index for evaluating the reliability of the package. Based on the fatigue life formula proposed by A.Syed, the fatigue life of the Flip Chip on Board is accordingly predicted.
Secondly, in order to analyze the effect of each control factor on the average strain energy density, the one-factor-at-a-time analysis is conducted. Afterwards, the factors with more significant effect are chosen as design parameters by the fractional factorial design method to set up the regression model of response surface. Subsequently, the genetic algorithm combined with the response surface method is applied to obtain the optimal design.
Finally, by the interval genetic algorithm, the sensitivity of each parameter can be tracked out, which are ranked from the largest to the smallest as follows：Coefficient of thermal expansion of underfill, diameter of copper column, thickness of printed circuit board and height of copper column. Thus it is also constrained by the manufacturing standards. Therefore it is urgent to define the range of the parameters under the constraints of cost and technology to ensure the response value of each parameter combination not beyond the allowable error range so that stability of the product quality can be expected.

[1]Patra, S.K., Lee, Y. C. “Quasi-Static Modeling of the Self-Alignment Mechanism in Flip-Chip Soldering-Part I:Single SolderJoint ”, Journal of Electronic Packaging, 113, pp. 337-342, 1991
[2]Brakke, K. “ The Surface Evolver ”, Experimental Mathematics,no. 2,pp. 141-165, 1992
[3]Lau, J.H. “Thermal Fatigue Life Prediction of Flip Chip Solder Joints by Fracture Mechanics Method”, Engineering Fracture Mechanics, Vol. 45, No. 5, pp. 643-&, 1993.
[4]Darveaux, R., Banerji, K., Mawer, A. and Dody, G. “ Reliability of Plastic Ball Grid Array Assembly ”, Ball Grid Array Technology, McGmw-Hill, pp. 379-442, 1995
[5]Lau, J.H. “Flip Chip Techinologies”, Mcgraw-Hill,New York,1996
[6]Zheng fang, Qian., Minfu, Lu., Wei Ren. and Sheng, Liu. “Fatigue Life Prediction of Flip-Chip in Terms of Nonlinear Behaviors of solder and Underfill”, Electronic Components and Technology conference, San Diego, CA, pp. 141-148, 1999.
[7]Wang, C.H., Holmes, A.S. and Gao, S. “Laser-assisted Bump Transfer for Flip Chip Assembly”, Electronic Materials and Packaging, Hong Kong, pp. 86-90, 2000.
[8]Syed, A. “ Accumulated creep strain and energy density basedthermal fatigue life prediction models for SnAgCu solder joints ”, Proceedings of the 54th Electronic Components and Technolog Conference, Las Vegas, Necada, USA, Vol. 1, pp. 737-746, 2004
[9]Pang J. H. L.and Che F. X. Thermal Fatigue Reliability Analysis for PBGA with Sn-3.8Ag-0.7Cu Solder Joints. In：Proceedings of 6th IEEE Electronics Packaging Technology Conference, pp.787-792,2004
[10]Stoyanov, S., Bailey, C. and Desmulliez, M. “ Optimisation modeling for thermal fatigue reliability of lead-free interconnects in flip-chip packaging ”, Soldering & Surafce Mount Technology, Vol. 21, No.1, pp. 11-24, 2009
[11]Dongliang Wang, Yuan Yuan, Le Luo, “ Failure Analysis ofSn-3.5Ag Solder Joints for FCOB Using 2-D FEA Model ”, Shang Hai Institute of Microsystem and Information Technology, Chinese Academy ofSciences, 2010
[12]Yong-Sung Park, Yong-Min Kwon, Jeong-Tak Moon, Young-Woo Lee, Jae-Hong Lee, and Kyung-wook Paik, “Effects of Fine Size Lead-Free Solder Ball on the Interfacial Reactions and Joint eliability ”, Korea Advanced Institute of Science and Technology (KAIST),2010
[13]Bagley J.D. The Behavior of Adaptive Systems Which Employ Genetic and Correlative Algorithms.
In：PhD thesis, University of Michigan, Ann Arbor, 1967.
[14]Shiau T.N., Kang C.H., Liu D.S. Interval optimization of rotor-bearing systems with dynamic behavior constraints using an interval genetic algorithm.
In：Structural and Multidisciplinary Optimization, Vol.36(6) ,pp 623-631, 2008

[15]羅盛沐「以區間式遺傳演算法進行堆疊晶QFN構裝體疲勞壽命之區間最佳化設計」，國立成功大學工程科學系碩士畢業論文，2009年
[16]潘嘉偉「探討關鍵錫球所累積平均應變能密度之反應曲面於 FCOB覆晶式載板接合封裝體之差異分析」，國立成功大學工程科學系碩士畢業論文，2010年
[17]John H. Lau, C.P. Wong, John L. Prince,Wataru Nakayama, “Electronic Package：Design, Material, Process, and Reliability”, The McGraw-Hill Companies, Inc.
[18]Irving H.Shames, Francis A. Cozzarelli, “Elastic and Inelastic Stress Analysis”,
[19]李輝煌「田口方法：品質設計的原理與實務」，高立圖書有限公司，2008
[20]葉怡成「實驗計劃法－製程與產品最佳化」，五南圖書有限公司，2001
[21]周鵬程「遺傳演算法原理與應用－活用Matlab」，全華圖書股份有限公司，2007
[22]萬政憲 「在熱循環作用下錫球結構與配置方式對PBGA構裝之可靠度探討」，成功大學工程科學系碩士畢業論文，1999年
[23]劉振中「無鉛錫球含多層金屬薄膜之晶圓級封裝結構應力分析”，成功大學工程科學系碩士畢業論文，2003年
[24]D. C. Montgomery “Design and Analysis of Experiments 6th edition” John Wiley & Sons, 2005
[25]Robert Kay, ”Advanced Microsystems Assembly using Screen Printing Technology”, IMAPS-UK MicroTech conference