Optimization of stir-squeeze casting parameters of SiC reinforced aluminum composite through Taguchi method

Abstract

Abstract Aluminum Matrix Composites (AMCs) are widely used in automotive applications due to their low density and superior mechanical properties. However, conventional stir casting often results in poor particle distribution and high porosity, while squeeze casting alone may not ensure adequate reinforcement dispersion. Therefore, optimization of the combined stir-squeeze casting process is required. The novelty of this study lies in the systematic optimization of stir-squeeze casting parameters for Al2024/SiC/Mg composites using the Taguchi method, with an emphasis on correlating processing parameters with porosity reduction, hardness enhancement, and tensile behavior through process-structure–property relationships. The composite was fabricated at 720 °C using stirring speeds of 150–500 rpm, stirring times of 5–15 min, and squeeze pressures of 0–025 MPa based on an L9 orthogonal array. Density, porosity, hardness, and tensile strength were evaluated, and signal-to-noise ratios were analyzed using the smaller-is-better criterion for porosity and the larger-is-better criterion for hardness and tensile strength. The results indicate that squeeze pressure is the most influential parameter in reducing porosity and increasing hardness, achieving a minimum porosity of 0.141% (reduced from 1.230% in the non-pressurized baseline) and a maximum hardness of 127.1 HV (improved from 81.26 HV), obtained at 25 MPa squeeze pressure. In contrast, tensile strength is more sensitive to stirring time, highlighting the role of particle distribution and microstructural uniformity. Overall, the optimized stir-squeeze casting parameters significantly enhance the physical and mechanical properties of Al2024/SiC/Mg composites, demonstrating the effectiveness of this method for high-performance AMC production.

Access to Document

Other files and links

• Link to publication in Scopus
• Open Access Version Available

Fingerprint