Harnessing High-Loading CNT in Al7075 Nanocomposites by Hybrid Processing Route: Tribomechanical Performance and Machine Learning-Driven Optimization
摘要
Aluminum 7075 alloy is widely used in the aerospace and automotive sectors due to its high specific strength, but its low wear resistance offers a considerable barrier for wider use. This study investigates the development of aluminum 7075-based nanocomposites reinforced with high concentrations of carbon nanotubes (1.5–3 wt%) using a hybrid fabrication approach that integrates rotational ultrasonication-assisted stir casting and friction stir processing. Although the initial casting step resulted in some nanoparticle agglomeration and casting-related defects, X-ray diffraction analysis confirmed successful incorporation of carbon nanotubes into the aluminum matrix. Post-casting, the samples were subjected to friction stir processing at a tool rotation speed of 900 RPM and a traverse speed of 25 mm/min, followed by T6 heat treatment to further enhance homogeneity and structural refinement. The results of microstructural examinations employing transmission electron microscopy (TEM), optical microscopy (OM), scanning electron microscopy (SEM), and energy-dispersive spectroscopy (EDS) showed uniform reinforcement distribution and refined grains. The friction stir-processed nanocomposites showed improved tensile and microhardness, along with significant enhancement in wear resistance, when compared to unprocessed samples. Additionally, machine learning techniques, specifically random forest regression combined with grid-based hyperparameter tuning, were employed to model and predict key mechanical and wear properties. The model identified the two most important characteristics as friction stir processing and carbon nanotube content. These nanocomposites have great promise for use in the automotive, aerospace, and military for applications requiring low weight, high strength, and resistance to wear.