摘要:"Spalling” 一詞指的是在微電子或微機電銲點內,介金屬從界面剝離的現象。這種剝離的現象之前就曾在IBM之phased-in覆晶銲點內之PbSn銲料與Cu/Cu-Cr 薄膜之反應中被觀察到。在此系統中,spalling的發生是由於與Cu6Sn5有良好潤濕性的Cu層被消耗殆盡,使得Cu6Sn5與潤濕性較差的Cu-Cr層接觸。由於Cu6Sn5與Cu-Cr層的潤溼性不佳,導致Cu6Sn5產生剝離。最近,另一種介金屬大規模剝離(massive spalling)的現象,在許多銲點內發生,而成為極為困擾業界的一項問題,這些銲點系統包含錫銀銅(SnAgCu)銲料與鎳(Ni)進行銲接反應,錫鋅(SnZn)與銅(Cu),高鉛銲料(PbSn)與銅(Cu),及高鉛銲料(PbSn)與鎳(Ni)的銲接反應。這種massive spalling的現象與過去在PbSn銲料與Cu/Cu-Cr薄膜反應中所觀察到的一般spalling有兩點明顯差異。首先,在這些系統中,spalling的發生是在較大規模的範圍發生,其次是此時具備良好潤溼性之金屬層並未消耗完。由於上述的兩項特點,此一大規模剝離現象被計畫申請人命名為massive spalling,並被眾多學者所沿用。Massive spalling會造成銲點之強度隨著高溫儲藏而劣化,最後導致沿界面脆性斷裂。這問題對銲點的可靠度影響甚鉅,因此,必須深入了解其基本機制並進行防治。雖然此一現象已經在許多不同的銲料系統中普遍被觀察到,但此一獨特之現象迄今仍無一個完整的理論解釋,因此本專題計畫將對此一兼具高度學術與應用重要性之主題進行深入之探討。
計畫申請人對促成massive spalling的驅動力(driving force)曾經提出一套解釋,這一套解釋雖然使我們對促成massive spalling之驅動力的瞭解有所幫助,但此一套解釋是以熱力學之觀點對此現象作分析,然而對massive spalling動力學及動態現象之探討仍然很缺乏。例如我們不了解介金屬如何能以整層剝離的方式離開界面,這個剝離的動作是如何逐步完成的?銲料是如何進入到剝離後之界金屬與基材之間? 表面張力是否在這個動態現象中扮演任何角色?上述諸多問題亟待進一步之研究才能回答。此外我們對各個階段之massive spalling下(massive spalling 未發生前、剛要發生時、spalling一半時、完全spalling之後)之銲點強度並不了解。也就是說,我們仍不清楚何種massive spalling微結構是最不利於銲點之強度及可靠度。而本研究計畫之目標即在嘗試解答以上這些重要之疑問。本計畫目標預計於三年內完成,第一年之子目標是探討銲墊上Au層,對massive spalling的影響。關於Au對spalling的影響目前還不清楚,需要進一步的實驗做更深入的探討。本年度除了對實驗現象完整觀察與描述外,並將對其機制作深入探討。第二年之子目標是探討massive spalling動力學及動態現象。試圖解答介金屬如何能以整層剝離的方式離開界面?這個剝離的動作是如何逐步完成的?銲料是如何進入到剝離後之界金屬與基材之間?表面張力是否在這個動態現象扮演任何角色?第三年之子目標是探討各個階段massive spalling狀況下之銲點強度。也就是要找出最不利銲點可靠度之微結構。
Abstract: In solder/substrate reactions, spalling refers to the detachment of the intermetallic compound, the reaction product, from the interface during the reaction. A classical example is the spalling of Cu6Sn5 during the reaction between the PbSn solder and the Cu/Cu-Cr thin film metallization. Spalling here was caused by the exhaustion of the Cu film, which had a good wetting with Cu6Sn5, and the subsequent direct contact between Cu6Sn5 and the remaining Cu-Cr layer, which had a relatively poor wetting with Cu6Sn5. This poor wetting between Cu-Cr and Cu6Sn5 caused the molten solder to penetrate into the roots of the Cu6Sn5 grains, leading to the spalling of Cu6Sn5 grains into the molten solder. Recently, another type of spalling was observed in several solder/substrate systems, including SnAgCu soldered on Ni substrate, SnZn on Cu, high-Pb PbSn on Cu, and high-Pb PbSn on Ni. In these solder/substrate systems, the intermetallics detached themselves in a massive scale during reflow or aging even though the wetting layer had not been depleted. Such a phenomenon was referred to as the massive spalling by the Principle Investigator of this proposal. In short, the massive spalling differs from the regular spalling in two ways: the apparent difference in the scale of spalling and the fact that massive spalling can occur even when the wettable layer is intact. Despite the fact that massive spalling has been observed across several solder/substrate systems, a unified explanation to rationalize this behavior is still lacking. The major objective of this proposal is to investigate this rather peculiar but practically very important phenomenon.
This Principle Investigator had proposed an explanation for the driving force for the massive spalling. Nevertheless, it should be emphasized that the arguments presented there were purely thermodynamic in nature. Nothing is said about dynamically how massive spalling occurs. Such unknowns include, for example, (a) how can the entire intermetallic layer detach itself from the interface, (b) how does solder enter the region between the substrate and the detached compound, (c) what is the role of the interfacial energy, etc. We are also interested in the mechanical properties of the joints at different stages of massive spalling. In other words, we want to know what type of microstructure resulting from the massive spalling is the most problematic for the reliability of the solder joints. The subjects in the list above are important and extremely interesting areas worthy of more studies. In this proposed study, we plan to answer these questions in three years. In the first year, we would like to explore the role of the Au layer in massive spalling. In the second year, the detailed step-by-step mechanism will be investigated. In the third year, the mechanical properties of solder joints at different stages of massive spalling will be assessed with the goal of identifying the most unfavorable microstructure resulting from the massive spalling.