Ternary blended concrete synergy of mineral admixtures

Concrete that uses supplementary mineral admixtures offers a route to reduce clinker use while maintaining performance, yet the combined action of multiple admixtures in one binder remains uncertain. The potential for synergy among metakaolin (MK), fly ash (FA), and rice husk ash (RHA) has been emphasized in prior work as a means to enhance packing and pozzolanic reaction. The gap addressed here is the absence of a practical way to quantify the combined efficiency of MK-FA-RHA and to predict strength across a broad range of ternary blends. The objective is to evaluate ternary MK-FA-RHA concretes and to derive synergy/efficiency-based equations to predict compressive strength and to correlate it with split tensile and flexural strength. An M²⁵ mixture with water/binder 0.45 and 39 combinations (MK 6−8%, FA 5−15%, RHA 5−20%) was produced; slump, compressive strength (7, 28, 56 days), split tensile and flexural strength were measured using IS:516 specimens (150 mm cubes, 75×150 mm cylinders, 100×100×500 mm beams). Workability decreased with increasing fines: at MK 7%, the slump fell from 188 mm to ≤100 mm as FA and RHA rose, and reached 35 mm at MK 8%, FA 10%, and RHA 20%. Strength responses showed that 8% MK alone raised 28- and 56-day compressive strength to 37.24 and 41.76 MPa (vs 34.87 and 38.87 MPa for the control), while RHA ≥15% produced 15−30% lower 28-day strength; the best ternary blend across all mixes was 8% MK + 10% FA + 10% RHA. Regression-based equations that incorporated a synergy factor accurately reproduced compressive strength, with most errors within 0−10%, and yielded R2 values of 0.73−0.82. Companion correlations predicted split tensile and flexural strengths from compressive strength. These findings suggest that MK 8% with FA and RHA at 10% each balances clinker reduction and strength, although high RHA contents require rheology control to avoid consolidation-limited results. Future work is recommended on durability mechanisms and admixture optimization to extend the predictive framework.