Low-carbon, high-strength concrete with high-volume Icelandic volcanic pozzolans and silicomanganese slag: A response surface methodology approach to predictive strength design

The interaction and combined effects of emerging natural and industrial supplementary cementitious materials (SCMs) at high-volume cement-replacement levels remain insufficiently understood, particularly for developing low carbon concrete without compromising mechanical performance. Although high-volume SCM replacement offers a practical route to reducing the CO₂ footprint of cementitious materials, the performance optimization of high-volume binary and ternary SCM binder systems remains limited. This study develops and validates predictive models for the fresh and hardened properties of low-carbon high-strength concrete incorporating Icelandic volcanic pozzolanic material (VPI) and silicomanganese (SiMn) slag at a total cement replacement level of 50% by weight. Response Surface Methodology (RSM), based on a Box–Behnken design, was employed to evaluate the effects of VPI–SiMn ratio, water-to-binder ratio and superplasticiser dosage on compressive strength (7 and 28 days), splitting tensile strength, elastic modulus, and workability. ANOVA confirmed model significance, identifying the water-to-binder ratio as the dominant factor governing strength and the superplasticiser dosage as the factor influencing workability, while allowing flexibility in the proportions of both SCMs within the design window. Sensitivity analysis quantified the contributions of linear, interaction, and quadratic terms, enabling reliable predictions. Validation experiments showed strong agreement with model predictions with deviation below 7%. A binary system with 50% VPI achieved a 28-day cube compressive strength of 107 MP, while a ternary system with 40% VPI and 10% SiMn slag attained 103 MPa with desirable fresh properties. These findings demonstrate that sustainable high-strength concrete can be produced using VPI and SiMn slag at 50% cement replacement and establish a robust framework for its performance-based optimisation.