Mechanical performance and predictive tribological modeling of Al7075 composites reinforced with rice hull activated carbon

This work reports the development of a sustainable Al7075 metal matrix composite reinforced with bio-derived activated carbon (AC) obtained from rice hull agricultural waste. Unlike conventional reinforcements such as SiC and Al₂O₃, rice hull-derived AC provides an eco-friendly, lightweight, and cost-effective alternative. The composites were fabricated using ultrasonic stir casting with varying AC contents (2-8 wt%). Microstructural characterization (OM, FESEM-EDS, and XRD) confirmed uniform dispersion of AC and the absence of detrimental Al₄C₃ formation. Mechanical testing revealed that 2 wt% AC yielded the optimum properties, improving hardness (by 21%) and tensile strength (by 23%) compared to unreinforced Al7075. Abrasive wear studies showed enhanced wear resistance and reduced coefficient of friction at the same reinforcement level. Beyond mechanical and tribological assessment, this work introduces a predictive framework using machine learning models (Gradient Boosted Trees, Gaussian Process Regression), which achieved near-perfect accuracy (R² > 0.99 for wear, R² > 0.96 for COF). These findings establish rice hull-derived activated carbon as a viable reinforcement for Al7075 composites and highlight the potential of data-driven approaches in predicting tribological performance, thereby advancing sustainable and intelligent material design.