J Fail. Anal and Preven.(2013)13: 194-201 Fig. 10 The relation between grain size and grain boundary area. (a) Fine grain thus more grain boundary area, and (b) coarse grain thus less grain boundary area pad was different. Normal pad displayed a bump sur inadequacy of energy to melt the surface layer exhib- face and failed pad displayed total flatness. The failed ted itself as un-wetting failure mode pad also revealed coarse and big grain size and even The pure tin solder used as confirm by EDS analysis on crystalline interface cracking under high magnification ormal pad was inappropriate. Since pure tin has lower The EDS analysis on normal pad assured us that the surface energy and higher melting point compare to solder used was pure tin and there was observable prevailing solders quantity of organic contamination during depository In the chemical reaction of enig. one nickel atom was replaced by two gold atoms. This fact combined with another fact that there is 16 difference in atomic uggestion radius between nickel and gold led to the misalignment It is proposed that the quality of the plating gold should of golden layer on nickel layer, forming a rugged be improved by adjusting the plating temperature, pH surface full of pits and holes. Furthermore, the crystal value, solution composition in the plating bath and the line interface cracking confirmed by SEM formed stirring mode. Thus, refining the grain size and sup- Grand Canyon"from a microscopic view. All these pressing the crystalline interface cracking facts described above exposed nickel to air and brought Replacing the pure tin with new solders like Sn-Ag-Cu about the formation of nickel oxide which further is strongly recommended acting as a barrier between solder and golden layer later during hot air solder leveling. Moreover, the " Grand Canyon"provided a channel for the upward diffusion References of nickel atom to the outermost surface and then be oxidized as a solder barrier. The"Grand Canyon"also Zeng, K Tu, K.N.: Six cases of reliability study of Pb-free solder absorbed organic contamination from surrounding joints in electronic packaging technology. Mater. Sci. Eng. 38(2),55-105(2002) environment into it playing a role as an obstacle for 2. Ji, L.N., Yang, Z.G., Liu, J.S. Failure analysis on blind vias of solder reaching nickel PCB for novel mobile phones. J. Fail. Anal. Prev. 8(6), 524-532 The coarse and big grain size revealed by SEM 3 aggravated the un-wetting. The big grain size corre Park, S.H., Yoo, J.S. Origin of surface defects in PCB final sponds to the reduction of total grain boundary area finishes by the electroless nickel immersion gold process. which possessed higher energy than that in bulk, thus J. Electron. Mater. 37(4),527-534(2008) lowering the energy of golden layer as a whole. So the 4. Chong, KM,Tamil,SS,Charan,G: Discoloration related failure mechanism and its root in electroless nickel immersion gold predetermined soldering energy from hot air solder (ENIG) pad metallurgical surface finish. In: Proceeding of llth leveling was not enough to fuse the golden layer. The IPFA,pp.229233(2004)pad was different. Normal pad displayed a bump sur￾face and failed pad displayed total flatness. The failed pad also revealed coarse and big grain size and even crystalline interface cracking under high magnification. The EDS analysis on normal pad assured us that the solder used was pure tin and there was observable quantity of organic contamination during depository. • In the chemical reaction of ENIG, one nickel atom was replaced by two gold atoms. This fact combined with another fact that there is 16% difference in atomic radius between nickel and gold led to the misalignment of golden layer on nickel layer, forming a rugged surface full of pits and holes. Furthermore, the crystal￾line interface cracking confirmed by SEM formed ‘‘Grand Canyon’’ from a microscopic view. All these facts described above exposed nickel to air and brought about the formation of nickel oxide which further acting as a barrier between solder and golden layer later during hot air solder leveling. Moreover, the ‘‘Grand Canyon’’ provided a channel for the upward diffusion of nickel atom to the outermost surface and then be oxidized as a solder barrier. The ‘‘Grand Canyon’’ also absorbed organic contamination from surrounding environment into it playing a role as an obstacle for solder reaching nickel. • The coarse and big grain size revealed by SEM aggravated the un-wetting. The big grain size corre￾sponds to the reduction of total grain boundary area which possessed higher energy than that in bulk, thus lowering the energy of golden layer as a whole. So the predetermined soldering energy from hot air solder leveling was not enough to fuse the golden layer. The inadequacy of energy to melt the surface layer exhib￾ited itself as un-wetting failure mode. • The pure tin solder used as confirm by EDS analysis on normal pad was inappropriate. Since pure tin has lower surface energy and higher melting point compare to prevailing solders. Suggestion • It is proposed that the quality of the plating gold should be improved by adjusting the plating temperature, pH value, solution composition in the plating bath and the stirring mode. Thus, refining the grain size and sup￾pressing the crystalline interface cracking. • Replacing the pure tin with new solders like Sn–Ag–Cu is strongly recommended. References 1. Zeng, K., Tu, K.N.: Six cases of reliability study of Pb-free solder joints in electronic packaging technology. Mater. Sci. Eng. R 38(2), 55–105 (2002) 2. Ji, L.N., Yang, Z.G., Liu, J.S.: Failure analysis on blind vias of PCB for novel mobile phones. J. Fail. Anal. Prev. 8(6), 524–532 (2008) 3. Kim, B.K., Lee, S.J., Kim, J.Y., Ji, K.Y., Yoon, Y.J., Kim, M.Y., Park, S.H., Yoo, J.S.: Origin of surface defects in PCB final finishes by the electroless nickel immersion gold process. J. Electron. Mater. 37(4), 527–534 (2008) 4. Chong, K.M., Tamil, S.S., Charan, G.: Discoloration related failure mechanism and its root cause in electroless nickel immersion gold (ENIG) pad metallurgical surface finish. In: Proceeding of 11th IPFA, pp. 229–233 (2004) Fig. 10 The relation between grain size and grain boundary area. (a) Fine grain thus more grain boundary area, and (b) coarse grain thus less grain boundary area 200 J Fail. Anal. and Preven. (2013) 13:194–201 123