AI-based machine learning prediction for optimization of copper coating process on graphite powder for green composite fabrication

Bearings are engineering components that must be engineered with high precision and quality to enable the machine components to rotate at high speeds. In the recent past Polytetrafluoroethylene (PTFE) has been used in the manufacturing of bearings since it has a low coefficient of friction. The present work uses graphite as filler in the PTFE matrix because of its excellent lubricant properties. However, polymer bearings cannot resist high temperatures. This deficiency can be overcome by coating graphite with copper and then using it as a filler material. Additionally, the use of copper-coated graphite as a filler material for green composite fabrication promotes sustainability by utilizing graphite, a naturally occurring mineral, and copper, a widely recyclable metal. An electroless coating technique was employed to get a uniform coating thickness of copper on graphite, following the coated graphite, which can be used as a filler material along with PTFE. The rate of deposition of copper on graphite particles depends on the sensitization time, activation time, and metallization time. In this work, a mathematical model integrated with a machine learning model is developed to predict the coating thickness eliminating the need to perform expensive experimental testing. The experimental design, guided by the Taguchi technique, incorporates the deployment of machine learning models. Specifically, a linear regression model with 76% accuracy and a Random Forest model with 96% accuracy is employed to automate and optimize the experimental process, ensuring efficient and precise results. The study holds potential for the fabrication of green PTFE composite materials.