透過您的圖書館登入
IP:18.222.163.31
  • 學位論文

銅濃度、體積及溫度的綜合效應對錫銀銅銲料與鎳基板間介面反應之研究

Combined Effects of Cu Concentration, Solder Volume, and Temperature on SnAgCu/Ni Interfacial Reaction

指導教授 : 高振宏
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


隨著電子產品愈講究多功能及輕薄短小,致使電子封裝技術也必須與時並進。而在電子封裝中,當銲點與基材進行銲接時,銲料會與基材上的金屬發生化學反應,進而在界面生成一或多種介金屬,而這些介金屬對銲點的可靠度影響甚鉅。因此,欲提升銲點可靠度,就必須充分了解銲點與基材間界面反應之情形。本研究的目的即是深入探討銲點組成、銲點體積以及溫度對於錫銀銅銲料與基板間之界面反應的影響。 五種不同銅濃度(0.3-0.7 wt.%)的錫銀銅銲料及三種不同大小的銲錫球(760,500及300微米)將與Ni基板進行銲接,以探討銅濃度及銲點體積的影響。此外,除了將試片經過90秒的迴銲之外,試片將持續放入160及180度C之烘箱中進行低溫熱處理,以討論在固-固反應以及不同溫度下之界面反應情形。 本研究發現,常見的錫銀銅銲料與鎳進行銲接時,共有(Cu,Ni)6Sn5與(Ni,Cu)3Sn4兩種介金屬於界面生成,有趣的是,此兩種介金屬的形態係與銲料中的銅濃度、銲點體積及反應溫度息息相關。在本實驗室早期的研究中發現,當含Cu的銲料與Ni反應時,些微改變銲料中的Cu濃度(0.6→0.4wt.%)會對界面反應有巨大的影響。當銲料中Cu濃度高於0.4 wt.%時,界面生成物為(Cu,Ni)6Sn5;當銲料中的Cu濃度低於0.4 wt.%,則界面生成物轉為(Ni,Cu)3Sn4。上述的結果是建立在bulk reaction的情形下,此時Cu的供應量接近無限大。然而在銲點中,由於銲料體積很小,再加上SnAgCu銲料之Cu濃度範圍一般低於1 wt.%。因此,在銲點中的Cu可供應的量是很有限的。當銲點與基材反應,生成含Cu介金屬時[如 (Cu,Ni)6Sn5],即可能造成銲點中之Cu濃度的大幅下降。 研究結果顯示,在液-固反應中,較大的銲錫球(760微米及500微米)與開孔375微米的Ni基板進行90秒銲接後,反應結果與之前bulk的反應結果是一致的。在Cu濃度高於0.4 wt.%時,生成的介金屬為(Cu,Ni)6Sn5;低於0.4 wt.%則生成(Ni,Cu)3Sn4。值得注意的是,當銲球的直徑縮小至300微米時,由於體積的限制使得銲點中的Cu含量相對變少,當界面上生成含Cu介金屬時,銲錫球中的銅含量會大量被消耗。在未反應前,銲錫球中的Cu濃度若高於0.4 wt.%則生成於界面的穩定相為(Cu,Ni)6Sn5。由於(Cu,Ni)6Sn5的生成,將造成銲錫球中的Cu濃度大幅下降。當銲料中的Cu濃度由高於0.4 wt.%轉變為低於0.4 wt.%時,原本穩定存在於界面的(Cu,Ni)6Sn5將不再穩定,因此界面將有一股由(Cu,Ni)6Sn5轉換成(Ni,Cu)3Sn4的驅動力,使得原本界面生成的(Cu,Ni)6Sn5藉由大規模剝離的方式離開界面,促使界面生成更穩定的(Ni,Cu)3Sn4。這種大規模剝離的現象,除了在SnAgCu/Ni的系統中觀察到之外,在SnZn/Cu、Pn5Sn/Cu及Pb5Sn/Ni系統中也都被觀察到。本研究也使用熱力學之觀點對四個不同的系統都會發生的大規模剝離之現象提出一個合理的解釋。界面介金屬會發生大規模剝離是因為隨著反應的進行,原本生成於界面的介金屬不再於銲料達成平衡,所以此介金屬就會藉由spalling的方式離開界面,使得新的介金屬可以在界面成核及成長。 本研究亦進一步發現,此種大規模的剝離現象並未出現在SnAgCu/Ni的固態反應中。在固態反應中,當界面的穩定平衡相,隨著銲料中Cu含量的減少而由(Cu,Ni)6Sn5轉變為(Ni,Cu)3Sn4時,界面穩定生成物的轉換將藉由交互擴散的方式將生成物由(Cu,Ni)6Sn5轉變成(Ni,Cu)3Sn4。除此之外,界面生成物由(Cu,Ni)6Sn5轉變成(Ni,Cu)3Sn4的臨界銅濃度也將隨著溫度的增加而上升。此溫度效應的結果,也與兩組研究團隊經由熱力學計算的結果相符合。因此本研究也將使用熱力學之觀點對此銅濃度、體積及溫度之效應做深入的探討。

並列摘要


Soldering is the most important joining technology in the electronic industry and the Sn-Ag-Cu serious of lead-free solders, around the Sn(3.5Ag±0.3)Ag(0.9±0.2)Cu (wt.%) ternary eutectic composition, are the most promising alternates. With the miniaturization of modern electronic products and devices, the packaging density of components such as chips significantly increases, and the scale of interconnections becomes smaller and smaller. Consequently, the reliability of solder joints plays key roles. And understanding the volume effect of SAC solder on interfacial reaction will be critical in solder joint reliability. Furthermore, temperature is also an important factor in the reaction between solder and substrate. In this study, the effects of Cu concentration, solder volume and temperature are studies concurrently. In this study, three different sizes of solder spheres (300, 500, and 760 micrometer diameter) with different Cu content (0.3 - 0.7 wt.%) were used to investigate the effect of Cu concentration and solder volume on interfacial reaction between SnAgCu and Ni. The samples were reflowed and subsequent aged at 160oC and 180oC. In our previous studies, the “Cu concentration effect” reported in the literatures were established in bulk reactions where the supply of Cu approached infinite. As a result, the Cu concentration could be assumed to be constant during the study. In those studies, the interface had reached local equilibrium and the Cu-Ni-Sn isotherm could be used to rationalize the formation of the reaction product(s). For example, when the Cu concentration was high, (Cu,Ni)6Sn5 was predicted to form next to the (Sn) phase. When the Cu concentration was low, (Ni,Cu)3Sn4 was predicted to form next to the (Sn) phase. When the Cu concentration was in-between, the diffusion path would pass through the three-phase field of (Sn) + (Cu,Ni)6Sn5 + (Ni,Cu)3Sn4 first. This study reveals both the Cu concentration and the solder volume had a strong effect on the type of the reaction products formed. The well-known “Cu concentration effect” refers to the sensitivity on the Cu concentration for the reactions between Ni and Cu-bearing solders. In a solder joint, the supply of Cu is limited because Cu is a minor constituent in solders. During the reaction, Cu in solder is incorporated into the reaction product(s), and as the amount of the product(s) increase, Cu in solder is gradually consumed. Consequently, the Cu concentration in solder might not be a constant during soldering. This change in Cu concentration may then makes the reaction product at the interface change from one compound to another. In the case that the Cu concentration does decrease from a concentration higher than 0.4 wt.% to less than 0.4 wt.% during reflow, the (Cu,Ni)6Sn5 which is original at the interface will spall massively because now Ni prefers to be in contact with (Ni,Cu)3Sn4. This massive spalling phenomenon is also observed in the other three systems, SnZn/Cu, high-Pb PbSn/Cu, and high-Pb PbSn/Ni and a unified thermodynamic argument is presented to rationalize the occurrence of the massive spalling in four different solder/substrate systems. The massive spalling occurs because during reaction the original reaction product at the interface is no longer in local thermodynamic equilibrium with the solder, and this compound is driving away to make room for the nucleation and growth of the equilibrium phase. Two necessary conditions for the occurrence of massive spalling are identified. The number one condition is that the reactive constituent must be present in a limited amount, and the second is that the soldering reaction must be very sensitive to the concentration of this element. However, in SnAgCu/Ni system, massive spalling phenomenon didn’t occur on solid-solid reaction as the equilibrium at interface changed. On solid-state reaction, (Ni,Cu)3Sn4 grows and (Cu,Ni)6Sn5 shrinks as the equilibrium compound changes from (Cu,Ni)6Sn5 to (Ni,Cu)3Sn4. Finally, the critical Cu concentration was a strong function of temperature. The relationship between critical Cu concentration and temperature has been calculated by Yu et al. [58] and Chen et al. [40]. Our study confirmed the critical Cu concentration for the formation of (Cu,Ni)6Sn5 or (Ni,Cu)3Sn4 decreases as temperature drops.

參考文獻


[63] C. W. Chang, S. C. Yang, C. T. Tu, and C. R. Kao, J. Electron. Mater., vol. 36, p.1455, 2007.
[49] C. C. Chang, Y.W. Lin, Y. W. Wang and C. R. Kao, J. Alloys Compd., vol. 492, p.99, 2010.
[1] P. Cusack, and T. Perr, “The EU RoHS Directive and its implications for the plastics industry,” Plastics, Additives and Compounding, vol. 8, p. 46, 2006.
[2] IPC Roadmap for Lead-Free Electronics Assemblies, 2nd draft, IPC, Northbrook, IL, November, 1999.
[3] National Electronics Manufacturing Initiative (NEMI) Lead-Free Readiness Task Force Report, NEMI, 1999.

延伸閱讀