Study On Process Optimization Of Polymer Surface Metallization By Using Artificial Neural Network

Materials design aims to manufacture materials with certain properties based on accumulated experience and scientific theory. Materials research has used a large number of experiments to determine the chemical composition and manufacture processes of materials, which lead to consume manpower, substance sources and time. By means of computer technology, we can obtain ideal materials by making use of fewer experiments. It is proved that the method of computer aided materials design is effective.Polymer surface metallization belongs to materials design, this thesis studies a new method of polymer surface metallization—chemical deoxidization. The detailed operation performed as follows: ABS resin was dipped in DMF solvent for 24 hours, mixed into some metallic salt and churned up sufficiently at 40℃for 4 hours, and then the flow-spread method was used to cast the mixed solution into mould. After drying at room temperature, the film was peeled from the mould easily, the area of the film was 3cm×1cm. New metalized films were obtained by the reduction of ABS/metal salt chelate films with NaBH₄ aqueous solution. The metalized films will exhibit different surface conductivity by changing the proportion of materials and the conditions of reduction reaction.This thesis gives an application of ANN approaches in polymer surface metallization for the sake of systemic research. Artificial neural network plays a leading role in the sciences for complex non-linear phenomena. It is suitable to resolve complex causal problems. The samples are arranged by uniform design method, based on experimental data, a data processing system named "PSM-Pb-Cu-SR" is programmed, which is based on the back propagation algorithm. ANN model is built by this program, which reflects the relationship between the surface conductivity properties of materials and its composition, the conditions of reduction reaction. A set of good agreements are obtained between experimental data and calculated data. It is found that the conditions of preparation influence on the surface resistance of the metal film, including the mass of ABS resin, the mass of metal salt, the temperature of reduction reaction, the concentration of the reducing agent and the time of reduction reaction. The best conditions of preparation are predicted by means of artificial neural network technology. It is advantageous to predict materials design with artificial neural network.