本文選題：體積收縮效應 + IMC垂直生長�。� 參考：《華中科技大學》2015年博士論文

【摘要】：Sn基焊料在電子器件中被廣泛地用于連接器件與基板,以提供所需的機械支撐和電信號通路�？煽炕ミB焊點的性能主要取決于在回流中熔化的焊料與固態(tài)基板間的反應。此反應在焊料/基板界面所生成的IMC層會在后續(xù)的器件服役期間或者高溫老化期間持續(xù)生長,導致焊點中的IMC比例不斷上升。由于IMC層(如Cu6Sn5和Cu3Sn)的力學性能,例如楊氏模量、硬度等,與焊點的焊料基體和被連接的金屬表面(如基板焊盤、引腳)有很大差別。因此,焊點中IMC比例的上升會導致焊點在外載荷作用下內(nèi)部變形明顯不協(xié)調(diào),這會引起重要的互連焊點可靠性問題。另外,IMC生長過程中同時伴隨著體積收縮。在微小焊點中,由于IMC的比例較高,IMC生長引起的體積收縮會進 一步導致焊點的尺寸變化、殘余應力、影響焊點的宏觀變形行為以及宏觀斷裂行為。在過去的幾十年中,雖然有大量的關(guān)于界面IMC層以及其對焊點可靠性影響的研究工作。但是,關(guān)于由IMC層生長導致的體積收縮所引起的焊點各種可靠性問題的研究,文獻中少有報導。 本文研究了Sn基無鉛焊點在老化后的IMC生長以及由IMC生長導致的體積收縮和由于體積收縮導致的焊點界面殘余應力的演化。另外,還對IMC層生長伴隨的體積收縮對焊點整體的壓蠕變變形和拉伸斷裂行為的影響進行了研究和分析。主要研究成果如下： 研究了Sn-1%Cu/Cu焊點中界面處沿擴散方向以及垂直于擴散方向的IMC生長。在175℃C老化之后,Cu6Sn5、Cu3Sn層的水平生長與老化時間的關(guān)系為h1MC=0.27t1/2+4.6,hcu6Sn5=0.16t1/2+4.1和hCu3Sn=0.17t1/2,其中t為老化時間,單位為h；h為IMC層厚度,單位為gm。在相同老化溫度下,垂直IMC的高度與老化時間呈拋物線規(guī)律,其關(guān)系式為y=0.11(?)t,其中t為老化時間,單位為h,y為垂直IMC的高度,單位為gm。 另外,垂直IMC為兩層結(jié)構(gòu),外層為Cu6Sn5內(nèi)部為Cu3Sn,但是CU3Sn先于Cu6Sn5形成。因此,IMC的垂直生長中Cu6Sn5和Cu3Sn的形成順序與水平IMC層中兩種IMC的形成順序相反。老化過程中,IMC沿兩個方向的生長與老化時間均呈拋物線關(guān)系,這表明兩個方向的生長均是由擴散主導的。但是,由于相對較長的擴散距離,IMC垂直生長速率明顯低于水平生長速率 本文還對由界面IMC層生長引起的焊點體積收縮進行了研究。在不同的老化時間之后,使用納米壓痕系統(tǒng)對經(jīng)過特殊設(shè)計的試樣表面的形貌進行測量。通過測量發(fā)現(xiàn),在175℃老化1132.5h后,焊點的塌陷可以達到1.2μm。并且,焊點的塌陷與老化時間在本文的實驗條件下符合△h=-0.031×(?)的關(guān)系。由此得到的尺寸變化系數(shù)(焊點塌陷高度與IMC層的厚度比)為αexperiment=、0.114。假設(shè)Cu6Sn5和Cu3Sn均為各向同性并且致密的結(jié)構(gòu),焊點塌陷的理論值為△htotal=-(0.04-0.004x)(?)t(其中x為由Cu3Sn生成的Cu3Sn的體積占總體積的比例；t為老化時間,單位為h)。因此,尺寸變化系數(shù)的理論值為αideal=-0.147+0.0147x。通過比較可以得出,焊點塌陷的理論計算結(jié)果與實驗測得的結(jié)果相符合。 焊點中IMC層生長導致的體積收縮極有可能受限于相鄰的焊料和基板,這可能導致焊料/Cu界面殘余應力的產(chǎn)生。因此,老化后,在Sn-1%Cu焊料、界面Cu6Sn5層、界面Cu3Sn層以及Cu基板進行控制深度的納米壓痕實驗,以研究焊點中殘余應力隨老化時間的演變。實驗結(jié)果顯示焊料/Cu界面不同部分的應力狀態(tài)與實際位置以及成分有關(guān)：(1)老化過程中,Cu3Sn/Cu界面附近的Cu基板中央以及邊緣均處于壓應力狀態(tài),其平均值為560MPa；(2)焊料/Cu6Sn5界面附近的焊料中央部分和邊緣的應力為70MPa和90MPa；(3)老化后,界面Cu6Sn5層中的壓應力在中央和邊緣分別上升至4GPa和3GPa；(4)相對地,在老化過程中,Cu3Sn層內(nèi)在中央和邊緣的拉伸應力均有上升,分別達到了1.7GPa和0.5GPa。 從焊點中相鄰兩部分的應力-老化時間(S-t)曲線的對比結(jié)果中,可以得出：(1)界面Cu6Sn5層和相鄰的焊料得到的S-t曲線沒有明顯的關(guān)聯(lián)；(2)在焊點中央,Cu基板和界面CU3Sn層內(nèi)的應力隨著老化時間呈相反的趨勢演變,這表明焊點中央的Cu3Sn層主要受相鄰的Cu基板的約束；(3)焊點邊緣的Cu3Sn層主要由相鄰的Cu3Sn層約束。 通過搭建蠕變實驗裝置,對SAC305/Cu焊點的壓蠕變行為進行了初步研究。通過研究發(fā)現(xiàn),當壓應力為17.8～22.9MPa、溫度為433～463K時,SAC305/Cu焊點的穩(wěn)態(tài)蠕變應變?yōu)?.6～1.4%,對應的穩(wěn)態(tài)壓蠕變變形為2.5～5.8μm。在該條件下Sn-1%Cu/Cu焊點穩(wěn)態(tài)蠕變過程中,界面處IMC層的厚度增加約1μm左右。由于IMC生長引起的高度變化為0.05～0.16μm,占穩(wěn)態(tài)壓蠕變過程中焊點高度下降總量的1～-4%。因此,在焊點穩(wěn)態(tài)蠕變過程中,IMC生長伴隨的體積收縮所引起的焊點高度變化對實驗測量得到的焊點蠕變變形總量的影響可以忽略不計。 通過在FIB制備的Cu6Sn5和Cu3Sn微柱上進行微懸臂梁測試,研究了Sn-1%Cu/Cu界面處的Cu6Sn5層和Cu3Sn層的拉伸斷裂行為。在微懸臂梁測試中,Cu6Sn和Cu3Sn微柱在外加載荷條件下一直保持彈性變形,直至斷裂。測試后,由Cu6Sn5微柱的斷口形貌可知Cu6Sn5的主要斷裂模式為穿晶斷裂和沿晶斷裂。而Cu3Sn微柱的斷裂僅存在沿晶斷裂。IMC微柱斷裂模式的不同主要是由界面Cu6Sn5層和Cu3Sn層的微觀組織結(jié)構(gòu)決定的。 另外,通過Cu/SAC305/Cu焊點的拉伸實驗,發(fā)現(xiàn)在老化前焊點的斷裂模式為韌性斷裂,斷裂強度為70.8±9.0MPa。當IMC顆粒沿焊料晶界處集中時,由于IMC生長伴隨的體積收縮會引起應力和缺陷集中,導致焊點的拉伸強度下降至58.9±6.8MPa。當焊點老化81h后,焊點從IMC層中斷裂,并且斷裂強度下降明顯,為57.8±8.4MPa。Cu/SAC305/Cu焊點的斷裂模式和斷裂強度的變化是由于IMC顆粒的特殊分布或者IMC生長伴隨的體積收縮導致IMC層中應力和孔洞的積累導致的。通過對比,發(fā)現(xiàn)IMC微柱斷裂強度遠高于焊點在IMC層中斷裂時的斷裂強度。這證明IMC生長過程中伴隨的體積收縮所導致的應力集中和孔洞等缺陷可以明顯降低焊點整體的可靠性。
[Abstract]:Sn based solders in electronic devices are widely used for connecting devices and substrate to provide the required mechanical support and electrical signal pathway. Reliable performance of solder joints mainly depends on the melting in the reflow solder in solid state reaction between the substrate and the reaction at the solder / IMC interface layer substrate generated will be served in the follow-up period or device during high temperature aging continued growth, resulting in the solder joints of IMC. Due to the rising proportion of the IMC layer (such as Cu6Sn5 and Cu3Sn) on the mechanical properties, such as young's modulus, hardness, and solder matrix of solder joints and the metal surface are connected (such as substrate pads, pins) are very different. Therefore, the rising proportion of IMC in the solder joints will lead solder joints under external load internal deformation is obviously not harmonious, it will cause the solder joint reliability issues important. In addition, the IMC growth process is accompanied by volume contraction in the micro. In small solder joints, due to the higher proportion of IMC, the volume contraction caused by IMC growth will come in.
Step size change caused the solder joint, residual stress, deformation behavior of solder joints and the influence of macro fracture behavior. In the past few decades, although there are a lot of research work on the interface of IMC layer and its influence on the reliability of solder joints. However, the research on the reliability of various solder joint by IMC layer growth volume contraction induced by the cause, the literature reports are very few.
This paper studies the Sn based lead-free solder joints after aging IMC growth and shrinkage caused by IMC growth and the evolution of the interface of solder joint residual shrinkage caused by stress. In addition, effects of IMC layer growth with volume shrinkage on joint overall compressive creep deformation and fracture behavior were studied and analyzed the main research results are as follows:
The interface of Sn-1%Cu/Cu solder joints along the direction perpendicular to the direction of diffusion and diffusion of the growth of IMC. After Cu6Sn5, C 175 degrees of aging, the relationship between Cu3Sn level growth and aging time of h1MC=0.27t1/2+4.6, hcu6Sn5=0.16t1/2+4.1 and hCu3Sn=0.17t1/2, where t is the aging time, the unit is h; h is IMC thickness, unit is GM. at the same aging temperature, aging time and vertical height of IMC a parabolic law and its relationship to y=0.11 (?) t, where t is the aging time, the unit is h, y is the vertical height of IMC, the unit is GM.
In addition, the vertical IMC is divided into two layers, the outer layer is Cu6Sn5 internal Cu3Sn, but CU3Sn prior to Cu6Sn5 formation. Therefore, the formation of two kinds of IMC sequence and the level of IMC layer in the form of Cu6Sn5 and Cu3Sn in reverse order vertical growth of IMC. In the process of aging time, IMC along two directions of growth and aging were a parabolic relation, which means that the two direction of the growth is dominated by diffusion. However, due to the diffusion of a relatively long distance, vertical IMC growth rate was significantly lower than the level of growth rate
We also studied by interfacial IMC layer growth of solder volume contraction induced. After different aging time, the use of nano indentation system through the special design of the surface morphology of the sample were measured. By measuring, at 175 DEG C after aging for 1132.5h, the solder joint can reach 1.2 m. and collapse, collapse and aging time of solder joints under the experimental condition with h=-0.031 * (?). The relationship between the size of the coefficient of variation (joint collapse height and IMC layer thickness ratio) is a experiment=, 0.114. Cu6Sn5 and Cu3Sn are assumed isotropic and compact structure, the solder joint collapse of the theoretical value is htotal=- (delta 0.04-0.004x) (?) t (wherein X is Cu3Sn generated Cu3Sn volume accounted for a proportion of the total volume of T; as the aging time, the unit is h). Therefore, the size change coefficients for a ideal=-0.147 +0.0147x. by comparing It is concluded that the theoretical calculation results of the collapse of the solder joints are in accordance with the experimental results.
Solder volume shrinkage of IMC layer growth is likely to lead to limited to the adjacent solder and substrate, which may lead to residual solder /Cu interfacial stress. Therefore, after aging at Sn-1%Cu solder interface Cu6Sn5 layer, Cu3Sn layer and Cu substrate interface of nano indentation experiment control the depth of the evolution of the residual of the solder joint the stress with the aging time. Experimental results show that the solder /Cu interface in different parts of the stress state and the actual position and relevant components: (1) in the process of aging, Cu3Sn/Cu near the interface of the Cu substrate are central and edge compressive stress state, the average value of 560MPa; (2) the central part of the stress of solder near the solder /Cu6Sn5 interface and the edge of the 70MPa and 90MPa; (3) after aging, the interface of Cu6Sn5 layer in the compressive stress in the central and edge were increased to 4GPa and 3GPa (4); in contrast, in the aging process, the inner layer of Cu3Sn The tensile stresses at both the central and the edge are all rising, reaching 1.7GPa and 0.5GPa., respectively.
From the adjacent solder joint stress and aging time of the two part (S-t) the comparative result of the curves, we can conclude that: (1) S-t Cu6Sn5 curve of the interface layer and the adjacent solder has no obvious correlation; (2) in the central Cu substrate and the solder joint, the stress in the interface of CU3Sn layer with aging time show the opposite trend of evolution, which indicates that the Cu3Sn layer is mainly influenced by the central solder Cu substrate adjacent to the constraint; (3) the edge of the Cu3Sn solder layer is mainly composed of Cu3Sn layer adjacent constraints.
By building the creep experimental device on compressive creep behavior of SAC305/Cu solder joints were studied. Through the study found that, when the pressure is 17.8 ~ 22.9MPa, the temperature is 433 ~ 463K, the steady state creep strain of SAC305/Cu solder joint is 0.6 ~ 1.4%, the corresponding steady-state pressure creep deformation is 2.5 ~ 5.8 M. in the Sn-1%Cu/Cu under the condition of steady state creep process of solder joints at the interface, the thickness of the IMC layer increases about 1 m. Because of the change of height caused by growth of IMC is 0.05 ~ 0.16 m, accounting for the steady pressure during creep of the solder joint height decline of the total 1 ~ -4%. so in solder joint steady creep process, negligible solder solder creep the height change of volume shrinkage of IMC with the growth of the measured total deformation effect.
The preparation of FIB Cu6Sn5 and Cu3Sn micro column on micro cantilever beam test of Sn-1%Cu/Cu at the interface of the Cu6Sn5 layer and Cu3Sn layer. The tensile fracture behavior of micro cantilever beam test, Cu6Sn and Cu3Sn micro column in the external load conditions remain elastic deformation, until the fracture. After the test, the main the fracture mode of fracture morphology by Cu6Sn5 shows that Cu6Sn5 micro column is transgranular fracture and intergranular fracture. The fracture of Cu3Sn micro column only exists intergranular fracture.IMC micro column fracture mode is mainly determined by the different microstructure of interface of Cu6Sn5 layer and Cu3Sn layer.
In addition, the tensile test of Cu/SAC305/Cu solder joints, found in fracture mode before aging solder joint is ductile fracture, the fracture strength was 70.8 + 9.0MPa. when IMC particles along the grain boundaries when the concentration of solder, the volume shrinkage of IMC with the growth will cause concentration and defects, resulting in the tensile strength of the solder joint decreased to 58.9 + 6.8MPa. when the solder joint after aging for 81h, from the IMC layer in the solder joint fracture, and fracture strength decreased significantly, change of fracture mode and fracture strength of 57.8 + 8.4MPa.Cu/SAC305/Cu solder joint is due to volume shrinkage or special distribution of IMC particles with IMC growth leads to stress and leads to the accumulation of holes in the IMC layer. By contrast, IMC micro column the fracture strength is much higher than the fracture strength of the solder joint crack interruption in IMC layer. It is proved that the IMC growth process with the volume contraction caused by stress concentration and holes can be obviously Reduce the reliability of the solder joint as a whole.

【學位授予單位】：華中科技大學
【學位級別】：博士
【學位授予年份】：2015
【分類號】：TG407

【相似文獻】

相關(guān)期刊論文 前4條

1 關(guān)海若,王笑靜,董忠武,王世斌;如何科學合理地繪制罐區(qū)自動化操作界面圖[J];煉油技術(shù)與工程;2004年11期

2 屈進紅;曹小剛;李才明;薛典軍;;航空物探數(shù)據(jù)庫系統(tǒng)定制界面技術(shù)研究[J];內(nèi)蒙古石油化工;2007年04期

3 裴海祥 ,王立新,張壽祿,常可;爆炸復合板復合界面微觀組織分析[J];壓力容器;2002年11期

4 ;[J];;年期

相關(guān)會議論文 前2條

1 柏勁松;陳森華;;多重交匯界面捕捉的LS方法改進[A];中國工程物理研究院科技年報（2003）[C];2003年

2 范美如;翟志剛;司廷;羅喜勝;楊基明;;不同形狀界面的RM不穩(wěn)定性數(shù)值研究[A];第十四屆全國激波與激波管學術(shù)會議論文集（上冊）[C];2010年

相關(guān)重要報紙文章 前2條

1 ;怎樣獲得軟件的整個界面[N];電腦報;2002年

2 內(nèi)蒙古 潘軍杰;翻唱，，我自錄自編！[N];電腦報;2004年

相關(guān)博士學位論文 前2條

1 陳志文;Cu-Sn互連界面的尺寸效應及微力學特性[D];華中科技大學;2015年

2 李洋;基于乘務(wù)員特性的機車駕駛界面優(yōu)化設(shè)計研究[D];西南交通大學;2012年

相關(guān)碩士學位論文 前10條

1 王棟;老年人智能手機界面圖標的辨識度研究[D];華南理工大學;2015年

2 譚芳芳;當下移動媒體界面設(shè)計的發(fā)展及交互性新探[D];天津美術(shù)學院;2015年

3 王凡;從“少即是多”中探究移動端扁平化UI界面設(shè)計[D];廣東工業(yè)大學;2015年

4 王萌;基于人因特性的武器操控界面設(shè)計[D];北京理工大學;2015年

5 郭凌峰;長虹電子部品公司ERP-PPCO軟件系統(tǒng)改進的設(shè)計與實現(xiàn)[D];電子科技大學;2014年

6 陳瑾藝;移動醫(yī)療產(chǎn)品App界面設(shè)計研究[D];南昌大學;2015年

7 袁超;壓力型錨桿錨固界面的流變特性試驗研究[D];湖南科技大學;2015年

8 張燦燦;鄭州市東風渠濱河堤岸界面積極利用的途徑研究[D];河南農(nóng)業(yè)大學;2015年

9 趙新輝;基于符號學的手機界面設(shè)計表達研究[D];齊齊哈爾大學;2012年

10 劉淑純;界面圖標的視覺設(shè)計與應用研究[D];湖南師范大學;2014年

本文編號：1736469