現代人的日常生活與各式各樣的電子產品緊密結合；而電子產品在不斷地求新求變的演變下，我們對電子產品的要求則是趨於輕、薄、短、小。如何在市場激烈競爭的環境與時間壓力之下，生產高品質、小尺寸的電子產品，對其相關的封裝製程來說則是一大挑戰。而在封裝製程當中，產品的缺陷可能是因為其充填物的不均勻性所造成的；而充填物的不均勻性則是與充填物濃度分佈不均、溫度分佈與局部凝膠化等問題有關。為了能夠了解這些現象，一般都是利用CAE模擬；然而，大多數的CAE模擬都是假設材料在封裝過程前後都是均勻的。而這樣的假設會與真實的材料特性相去甚遠，容易忽略掉實際會發生的現象，例如剪切率誘發粒子遷移與粒子沉降等現象。
在本研究中，我們利用三維模擬模型來預測非膠體填充物懸浮液在微晶片封裝中填充物的濃度分佈。首先，我們所假設的模型先用二維通道和軸對稱圓管的幾何模型來驗證；結果顯示，填料濃度分佈的趨勢與實驗結果相符。除此之外，轉移模製的封裝製程中造成填料濃度的不均勻性有的因素甚多；因此在製程中，何種驅動力或原因造成成型後充填物的濃度分佈不均與分佈情形是其探討之重點，而目前學者常用來探討的成因有粒子沉降和剪切率誘發粒子遷移等因素。本研究則是著重在轉移模製的封裝製程中，並透過不同的條件：成型時間、模具溫度及塑料溫度，來探討封膠內充填物的濃度分佈情形。研究結果發現，成型時間對充填物的濃度分佈影響較為明顯，而我們也清楚地模擬填料濃度分佈在封膠後的整體濃度分佈不均，並探討了其分佈的趨勢。借由分析模擬結果，我們可以簡易預測在不同操作條件下封裝過程中填料的濃度分佈情形，並能有效地降低製造成本和產品設計週期。

We are using a wide range of electronic products in our daily life, and we can hardly live without them. Due to our requirements for modern electronic product - lighter, slimmer , shorter, and smaller, the smaller dimensional specification is expected to fit under the evolution of continuous process. On the other hand, under the highly competitive environment and the pressure of time-to-market, how to provide qualified modern electronic products is a new challenge for the associated package process. Specifically, one of the important defects is non-uniform property result from un-even filler particle distribution, temperature profile, and local gelation change during the package process.To catch the phenomena, CAE simulation is commonly applied. However, most CAE simulations assume the material property is homogeneous, and this assumption will be far from the situation in reality such as regarding to shear-induced migration and particle settling.
In this study, a three-dimensional simulation model of non-colloidal filler suspension is proposed to predict the filler concentration in microchips. Firstly, the proposed model is validated using two-dimensional channel and axisymmetrical circular pipe geometry model. Results showed that the trend of filler distribution is in a good agreement. Furthermore, there are various factors caused the inhomogeneity of fillers during the encapsulation of transfer molding processes. Therefore, it's important to figure out what the driving forces or causes are. Study shows there are two main reasons in the processes induce distinct filler concentration distributions after molding - settling and shear migration. We focus on the transfer molding processes in this study, and we discuss different conditions such as transfer time, mold temperrature, resin temperature, which affect the filler distribution. The results show that the transfer time may be the major factor than the others, and they will provide a comprehensive understanding uneven filler concentration after encapsulation. By using the integrated analysis, filler concentration under deferent working condition during encapsulation can be easily predicted, so as to efficiently reduce manufacturing cost and design cycle time.

[1] 鍾文仁, IC封裝製程與CAE應用, 全華圖書股份有限公司(2010).
[2] 許明哲, 先進微電子3D-IC構裝, 五南圖書出版股份有限公司(2011).
[3] 劉士榮, 高分子加工, 滄海書局(2010).
[4] 何宗漢, 半導體構裝製程簡介(2012).
[5] 田民波, 半導體電子元件構裝技術, 五南圖書出版股份有限公司(2005).
[6] 科盛科技股份有限公司編著, Moldex3D/Solid真實三維模流分析理論與應用, 科盛科技(2009).
[7] H. Yue, D. Bigio, M.G. Pecht, Components and Packaging Technologies, IEEE Transactions on 27 (2004) 493.
[8] W. Jinlin, T. Chen, The effect of flow properties on filler settling of underfill in the flip chip package, Electronic Components and Technology Conference, 2004. Proceedings. 54th(2004), p. 761.
[9] 林益生、許蓁容、何宗漢、伍玉真、鄧希哲, 工程與科技教育期刊 7 (2010) 532
[10] R.J. Phillips, R.C. Armstrong, R.A. Brown, A.L. Graham, J.R. Abbott, Physics of Fluids A 4 (1992) 30.
[11] 曾煥錩*、張元榕、許嘉翔, 最新3D-CAE模流分析金屬射出成型之粉膠分離現象, 2014 年模具暨應用產業技術論文發表會(2014).
[12] R.-y. Chang, W.-h. Yang, International Journal for Numerical Methods in Fluids 37 (2001) 125.
[13] J.F. Morris, F. Boulay, Journal of Rheology 43 (1999) 1213.
[14] R.M. Miller, J.F. Morris, Journal of Non-Newtonian Fluid Mechanics 135 (2006) 149.
[15] J. Morris, Rheol Acta 48 (2009) 909.