Optimization of the Sandblasting Process for a Better Electrodeposition of Copper Thin Films on Aluminum Substrate by Feedforward Neural Network

The influence of a proper surface preparation is essential for a better adhesion of copper thin films on aluminum substrate. In this work, the surface properties of the aluminum substrate have been modified through sandblasting process, in order to influence the quality of electroplating. To evaluate the correct adhesion of the thin film to the substrate non-destructive measurements of diffusivity by infrared thermography have been made. A combining of a feedforward artificial neural network (FFANN) and an external optimized algorithm (EOA) is proposed to optimize the substrate surface preparation process. A FFANN model is developed to map the complex non-linear relationship between the surface process conditions of the substrate and the thermal diffusivity of the electroplated sample. A good performance of the FFANN model is achieved. An EOA is used for the optimization of the sandblasting process conditions, in order to maximize the adhesion of the thin film to the substrate.