Tribological analysis and machine learning modeling of nickel-coated graphite reinforced A206 metal matrix composites

Abstract

This paper demonstrates the successful dispersion of up to 8 wt% nickel-coated graphite (NiGr) particles in A206 aluminum alloys using stir mixing followed by casting. The microstructure, as well as selected mechanical and tribological properties, were thoroughly characterized. The nickel coating improves the dispersion of graphite, overcoming the challenges faced with uncoated graphite in aluminum melts and ensuring uniform distribution of NiGr particles. The A206/NiGr composites show potential as replacements for aluminum-tin alloys, particularly in bearing applications. Wear and friction performance were evaluated using a pin-on-disc tribometer with a 440 stainless steel counterface. Both as-cast and worn surfaces were examined via optical microscopy, scanning electron microscopy (SEM), and energy dispersive spectroscopy (EDS). Additionally, a supervised machine learning algorithm was developed to model the relationship between NiGr content, heat treatment, load, and wear rate. The results indicate that the coefficient of friction (COF) decreases with NiGr additions up to 4 wt%, demonstrating self-lubricating behaviour. However, the wear rate increased with higher NiGr content due to reduced hardness. The machine learning model effectively predicts wear rates based on NiGr percentage, heat treatment, and load, highlighting NiGr content as the most significant factor influencing wear, followed by load and heat treatment.