Wear performance and optimization of h-BN-modified AZ91/TiB ₂ magnesium composites under dry sliding conditions

This study investigates the dry sliding tribological behavior of AZ91 magnesium alloy reinforced with titanium diboride (TiB 2 ) and hexagonal boron nitride (h-BN) ceramic particles. The primary objective was to enhance wear resistance and reduce the coefficient of friction (COF) through a synergistic reinforcement approach. Hybrid composites were fabricated using a two-stage stir casting process integrating mechanical and ultrasonic agitation to ensure uniform dispersion of the reinforcements. Experiments were conducted using a Taguchi L16 orthogonal array, and the influence of normal load, sliding speed, h-BN content, and sliding distance on wear and COF was evaluated under ASTM G99 conditions. Signal-to-noise ratio analysis revealed that normal load and h-BN content were the most significant factors influencing wear rate and COF, respectively. Gradient Boosting Regression (GBR) models were developed to predict the responses, achieving R 2 values of 0.816 for wear rate and 0.825 for COF. The GBR models were then integrated with a Multi-Objective Particle Swarm Optimization (MOPSO) algorithm to identify Pareto-optimal conditions. The optimum solution (Normal load = 40 N, Speed = 1.0076 m/s, h-BN = 1.1731wt.%, Distance = 1 km) yielded wear rate and COF values of 14.98 mg/km and 0.2517, with <4% deviation from predictions. SEM analysis of worn surfaces confirmed mechanisms including micro grooving, tribolayer formation, and debris-controlled wear. The GBR–MOPSO hybrid framework proved effective for optimizing and understanding the complex wear-friction interplay in AZ91 hybrid composites.