;Packaging wire bonding technology has been around for ages, is electrical connection technology, widely used in the microelectronics field, facing high density, complexity, small size of portable low-cost products, we need to implement mounting technology to match. General industry use material which is close to gold 4N (99.99%) as a package lead wire, but with the price of gold per ounce from USD 255 in 2001, rose to the highest in 2011 at USD 1900 per ounce, which result in manufacturing cost increase and profit goes down for the industry which use pure gold as main packaging wire.
In order to get rid of the sharp rise in the cost of material, packaging factory began to replace gold wire by copper wire. However, copper is easily oxidized and difficult to store, and in the wire bonding need to use an inert gas; on the other hand because of the higher strength and hardness of the copper wire make wire bonding operation parameters encountered narrower, slower and yield high defect rate, above also limits the copper wire in the electronics packaging industry popularity.
Although the price of silver is higher than copper, but lower than gold. Silver don’t has the nature of Copper easily oxidized, too high strength and hardness. In the packaging wire bonding process, Silver play the new role of wire material. During wire work, wire machine with lots of complicated parameters can be adjusted. In order to shorten the time to adjust parameters of wire machine, let silver wire and the baseboard has a better bonding, which can reduce the possibility of defect rate on following process.
In this study, 2N silver wire (99%) 0.8 mil diameter as the theme, after the wire annealing, cleaning, baking, referring Taguchi orthogonal arrays, using vacuum time and wire parameters as factors, after wire bonding, pointing at bondability between silver wire and base board, perform tensile test and push ball test, and the application with TOPSIS for experimental cooperation merits of multiple quality sorting, and then by gray correlation method for solving the optimal experimental parameters, using statistical analysis software within the SN ratio prediction function to predict and compare the results, verify and obtain the best process parameters.