Optimizing self-consolidating concrete with limestone filler by using statistical factorial design methods

Abstract

Self-consolidating concrete (SCC) is typically proportioned with relatively high contents of cementitious materials and chemical admixtures, leading to relatively high material cost. Such cost can be tolerated in high-value-added applications, especially when cost savings can be realized from using SCC, given the simplification of concrete placement and the reduction in construction time and labor cost. Efforts are still needed to reduce material cost for SCC to gain wider acceptance in a variety of applications. In addition to proper material selection and sound mixture proportioning, the incorporation of readily available fillers can enable reduction in cement and admixture contents, leading to savings in cost. This paper presents the results of an experimental program in which response surface methods are employed to optimize a four-component concrete containing limestone filler subject to eight performance criteria. These criteria include slump flow and its retention over time, V-funnel flow time, rheological parameters, surface settlement, and compressive strength at 1 and 28 days. The statistical response models are valid for mixtures made with water-to-powder ratios of 0.38 to 0.72 that contain up to 120 kg/m3 of limestone filler, 250 to 400 kg/m3 of cement, and a high-range water-reducing admixture (HRWR) dosage of 0.12 to 0.75% by mass of powder (cement and limestone filler). Fixed volumes of coarse aggregate and welan gum viscosity-enhancing admixture were employed to ensure adequate stability. The replacement of 100 kg/m3 of cement with finely ground limestone filler is shown to improve deformability and stability without affecting the one-day compressive strength. The concrete can exhibit up to 10% lower 28-day strength compared with similar concrete without filler. The use of filler reduces cost, as it enables the lowering of the cement factor and HRWR demand. Key properties of SCC, estimated from the derived models, are compared with values obtained from previously proposed models that were developed for structural SCC targeted for filling highly restricted sections. Although the binder composition differs for the two concretes, it is shown that the proposed response models still offer useful information regarding mixture optimization to securing cost-effective, high-performance SCC.