PCB007 Magazine


Issue link: https://iconnect007.uberflip.com/i/1185676

Contents of this Issue


Page 67 of 113

68 PCB007 MAGAZINE I NOVEMBER 2019 Au are purpose defined. Where comparisons are made, CN Au1 will be used on high-phos- phorous (HP; P > 10%) nickel and CN Au2 will be used on mid-phosphorous (MP; P = 7-10%) nickel. In essence, the CNF Au bath will be compared against specialist golds. At approximately 10.5% phosphorous con- tent in the electroless nickel, the crystal struc- ture of the layer changes from nano-crystalline to amorphous [1] . In doing so, the availability of atoms to promulgate the immersion gold process is reduced (Figure 3). This is an expla- nation as to why the corrosion performance is better for HP nickel but is also why specialty application-orientated immersion gold baths are required for products. CNF Au General Characteristics The initial goal will be to establish some ba- sic characteristics for the CNF Au bath. All of the basic characteristics were established by plating on an HP nickel. The basic characteris- tics will be judged according to age, time, tem- perature, and agitation. Impact of Metal Turnovers on Gold Deposition Cyanide is used as a key complexing agent specifically for gold and is in place to prevent plate-out. Cyanide alternatives are judged by their relative stability because instability is di- rectly related to cost; in the case of gold, that's a significant cost! Metal turnovers (MTOs) are an accepted life- time scaler in the electronics industry. MTOs refer to how many times the metal at make- up is added back through replenishment. For example, if the makeup of the bath is 1 g/l gold, 10 MTOs will be achieved when 10 g/l of gold has been added back to the bath. Figure 4 demonstrates that the gold thickness over a lifetime is statistically stable. While stability is a key attribute for immersion gold, cost con- cerns also require that the distribution of thick- ness be narrow. The coefficient of variance (CV%) is a rep- resentation of distribution. In this testing envi- ronment, a coefficient of less than 10% is con- sidered good. Figure 5 demonstrates that there is no statistical difference in CV% between the beginning of the lifetime and the end of the lifetime. For this article, the IPC 4552A meth- odology was not used, as the pad sizes that were measured varied significantly in size. Figure 3: Impact of phosphorous content on crystallinity [1] . Figure 4: ANOVA, the impact of metal turnovers on gold deposition. Figure 5: ANOVA, the coefficient of variance vs. MTO.

Articles in this issue

Archives of this issue

view archives of PCB007 Magazine - PCB007-Nov2019