1. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Mechanical and microstructural properties of SnAgCu solder joints 報 告 者 : 楊顯奕 指導教 授 : 林克默博士 日 期 :2011/03/29 Janne J. Sundelin a, ∗, Sami T. Nurmib, Toivo K. Lepist¨o a, Eero O. Ristolainen b a Institute of Materials Science, Tampere University of Technology, P.O. Box 589, FIN-33101 Tampere, Finland b Institute of Electronics, Tampere University of Technology, P.O. Box 692, FIN-33101 Tampere, Finland Received 1 March 2005; received in revised form 1 January 2006; accepted 1 January 2006

2. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Outilne Introduction Experimental Results Discussion Conclusuons 2

3. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Introduction Environmental and health concerns of lead have increased the pressure towards the lead-free soldering in electronics industry. Organic solderability preservative (OSP) surface finish is used on Cu pads mainly to prevent oxidation of Cu. Experimental studies have shown that OSP does not affect the metallurgy of the solder joints. Several SnAgCu compositions for replacement of traditional SnPb solder alloys have been proposed by different consortia; in Japan 96.5Sn3.0Ag0.5Cu, in US 95.5Sn3.9Ag0.6Cu, and in EU 95.5Sn3.8Ag0.7Cu. 3

4. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 The use of NiAu surface finish is generally argued by the fact that Ni barrier prevents the diffusion of Cu atoms into the solder and thus excessive growth of intermetallic compounds [1]. The role of a thin Au layer used on Ni is to improve wetting properties, and to protect the pads from corrosion. Found that both the cooling rate and the composition will affect the size of Ag3Sn dispersions in SnAgCu solder alloys, and this in turn was suggested to affect the mechanical properties of the alloys. 4

5. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Experimental Fig. 1. Schematic presentation of the single-overlap specimens used in the shear tests. 5

6. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Fig. 2. The reflow profiles for SnAgCu and SnPb solder joints. 6

7. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 The strength of the solder joints was measured in an universal Instron 4411 test machine in the as-reflowed condition and after thermal aging. Five tests were performed for each solder/surface finish combination in both the as-reflowed and aged conditions.The specimens were loaded and strained to failure at room temperature using a displacement rate of 2 mm/min. 7

8. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Cross-sectional samples were made after the shear tests from each solder/surface finish combination for microstructural characterization.The microstructure of the solder joint was analyzed with a scanning electron microscope (SEM) Philips XL30 and the compositional data was obtained with energy dispersive spectrometer (EDS) Edax DX4. Also, the paths of the fracture were studied with samples taken from the joints fractured in the pull test. 8

9. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 3. Results 3.1. Shear test results 9

10. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Fig. 3. The shear test results in the as-reflowed state and after isothermal aging:(a) on OSP finish and (b) on NiAu finish. 10

11. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Comparing the results of the joints with different PCB surface finishes, OSP and NiAu, it can be seen that in the as-reflowed condition the joints on OSP exhibited higher shear strength values than the ones on the NiAu finish. However, after the aging the difference is reduced significantly. This suggests, that the microstructure of the SnAgCu solder joints is different on OSP and NiAu, but during the thermal aging these differences in the microstructure are balanced. 11

12. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Fig. 4. The typical appearance of fracture surfaces. The fracture surfaces are from eutectic solder joints on: (a) OSP finish and (b) NiAu finish without aging. The fracture was ductile along the bulk solder. 12

13. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Fig. 5. The microstructure of bulk solders on OSP finish with: (a) eutectic and (b) hypoeutectic composition. The dendritic structure of Sn in eutectic solder is distinct. 13

14. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Discussion 4.1. Effect of PCB surface finish on the microstructure In the samples used in the present studies, the total area of pads was substantial, and therefore the effect of surface finish on the cooling rate in the reflow process is significant. The coarsening effect of slower cooling rate in SnAgCu microstructure has been verified, e.g. by Kim et al. 14

15. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Fig. 6. The dependence of the size of Ag3Sn dispersions from the PCB surface finish and SnAgCu composition. (a) Eutectic SnAgCu on OSP; (b) eutectic SnAgCu on NiAu; (c) hypoeutectic SnAgCu on OSP; (d) hypoeutectic SnAgCu on NiAu; (e) hypereutectic SnAgCu on OSP; (f) hypereutectic SnAgCu on NiAu. 15

16. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 4.2. Effect of SnAgCu composition on the microstructure Fig. 7. The intermetallic layers between SnAgCu and (a) OSP finish, Cu6Sn5, (b) NiAu finish (Cu,Ni) 6 Sn 5. 16

17. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Fig. 8. The average thickness of intermetallic layers measured from three different locations in the joint. 17

18. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Fig. 9. The coarsening effect of aging at 85 ◦ C for 1000 h on the Ag 3 Sn dispersion size in eutectic SnAgCu solder joints on: (a) OSP and (b) NiAu surface finish. 18

19. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Fig. 10. Sn-rich part of ternary SnAgCu phase diagram [12]. 19

20. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 Conclusions 在 PCB 板上，表面的光滑度，可能會去影響到銲點 的冷卻速度。在冷卻速度上， NiAu 的冷卻速度會比 OPS 慢，這是因為銅鎳的熱阻高於銅。 在老化實驗裡， OSP 抗點強度較低。 亞共晶，共晶和過共晶成分被發現有不同的微觀結 構，特別是在回流狀態。 在共晶錫銀銅分散的 Ag3Sn 明顯高於亞共晶錫的銀 銅，這是因為銀含量較高。在過共晶關節，大部分 的銀被消耗在一開始時的 Ag3Sn 。 20

21. 版權所有 翻印必究 2016/3/9 STUT 太陽能材料與模組實驗室 21 Thank you for your attention