Coupled Effects of Silica Fume and Steel-Fiber Geometry on the Mechanical and Elastic Performance of Concrete: Experimental Investigation, Statistical Interaction Analysis, and Predictive Modelling

Abstract

This study investigates the coupled effects of silica-fume replacement, steel-fibre dosage and fibre diameter on the mechanical, elastic and non-destructive performance of concrete through an integrated experimental and statistical framework. Silica fume was investigated at replacement levels of 4, 8 and 12%, with an additional 16% mixture used to establish the post-optimum response, while steel fibres were incorporated at 0.5, 1.0 and 1.5% using diameters of 0.5 and 1.0 mm. Performance was assessed through workability, cube and cylinder compressive strength, flexural strength, static and dynamic modulus of elasticity, ultrasonic pulse velocity (UPV), and compressive stress–strain behaviour, primarily at 28 and 90 days. Factorial analysis of variance and empirical regression were used to distinguish the relative contributions of the experimental variables and examine relationships among measured responses. Silica-fume incorporation increased cube compressive strength up to approximately 12% replacement, beyond which strength declined. The best-performing combination within the investigated domain contained 12% silica fume, 1.5% steel fibre and 1.0-mm-diameter fibres. At 90 days, this mixture achieved a cylinder compressive strength of 42.90 MPa, dynamic modulus of 48.30 GPa and UPV of 4752 m/s, corresponding to improvements of 20.34, 12.16 and 6.35%, respectively, relative to normal concrete. Its flexural strength reached 6.85 MPa, representing a 26.6% improvement over the corresponding reference concrete. Factorial ANOVA showed that silica-fume content exerted the dominant influence on compressive strength, whereas fibre dosage and fibre diameter had comparatively stronger effects on flexural performance. Higher-order interactions were not statistically significant, indicating that the observed improvements arose predominantly from cumulative main effects rather than a universal three-factor synergy. Representative empirical relationships produced coefficients of determination up to 0.977 within the investigated domain. The results demonstrate that multi-response assessment provides a more defensible basis for proportioning silica-fume steel-fibre concrete than compressive-strength optimisation alone, particularly where improvements in crack-sensitive behaviour must be balanced against increasing workability and admixture demand.