Modeling and Response Surfaces for Electroless Co Deposition Process

The electroless deposition process was studied by statistical techniques using IBM Storage Systems Division statistical computer program ADOE (Automatic Design of Experiments). Three types of experiments were disigned: the two-factor two-level factorial, the two-factor three-level factorial, and the central composite rotatable five-factor five-level experiments. The two-factor factorial experiments were used to determine the effect of each individual variable (Factor) and interactions between variables. It was concluded from the analysis of data that (i) the most significant factors are citrate, the Co2+complexing agent, pH, and temperature, and (ii) the most significant interactions are pH-citrate, pH-hypophosphite, and citrate-boric acid interactions. The central composite rotatable experiments were used to construct the second-order empirical model for the relationship between the Co deposition rate and five independent variables. The interpolation polynomial had 21 coefficients. The interpolation function was used to construct the 3-D response surfaces and 2-D response contours. Two types of response surfaces were found: simple maximum and rising ridge. The interpolation polynomial was used also to optimize the Co deposition process with respect to the minimized variability.