Effects of mineral admixtures on the fresh and hardened properties of self compacting concretes: Binary, ternary and quaternary systems

SCC is a new kind of high performance concrete with excellent deformability and segregation resistance. It can flow through and fill the gaps of reinforcement and corners of moulds without any need for vibration and compaction during the placing process. Use of SCC lowered the noise level on the construction site, diminished the effect on the environment, and increased productivity. Use of SCC, thus improved both the conditions for the labour on the work site and for the surroundings. In order to achieve a SCC of high fluidity and to prevent the segregation and bleeding during transportation and placing, the formulators have employed a high portland cement (PC) content and used superplasticizer, viscosity modifying admixtures and mineral admixtures. In this study, an experimental program was conducted to investigate the effects of mineral admixtures used in binary (two components), ternary (three components), and quaternary (four components) cementitious blends on the mechanical, physical, and durability properties of self compacting concretes. In this frame work four mineral admixtures, namely fly ash, ground granulated blast furnace slag, silica fume, and metakaolin were employed. A total of 65 self compacting concrete mixtures were designed at 0.32 and 0.44 water/binder ratios. The fresh properties of the produced self compacting concretes were observed through slump flow diameter, slump flow time, V-funnel flow time, L-box height ratio, initial and final setting times, and viscosity. The concretes were also tested for the mechanical, physical, and durability properties. The hardened concretes were tested for the compressive strength, splitting tensile strength, and ultrasonic pulse velocity for the evaluation of mechanical properties. Shrinkage accompanied by the water loss were also monitored. Moreover, the durability tests were conducted to investigate the resistance to chloride ion penetration, electrical resistivity, water absorption, sorptivity, and water permeability. Based on the test results, the fresh and hardened properties of self compacting concretes were predicted depending on the mixture constituents of SCCs using artificial neural network (ANN) and genetic programming (GEP) techniques. Using the explicit artificial neural networks based mathematical expressions single and multi objective optimization problems are formulated and solved to obtain the best possible values for ingredient of SCCs for single and several responses. The test results revealed that it is possible to produce SCC with binary, ternary and quaternary blends of fly ash, ground granulated blast furnace slag, silica fume and metakaolin. The binary use of fly ash and ground granulated blast furnace improved the fresh properties but silica fume and metakaolin had adverse effect on the fresh properties. Mechanical properties of produced SCCs improved with ground granulated blast furnace slag, silica fume and metakaolin but decreased with FA. It was observed that use of mineral admixtures as binary, ternary and quaternary blends improved the durability properties of SCCs. In addition, ANN appeared to have a better prediction performance than GEP by having owing to the higher coefficient of correlation and the lower mean absolute percentage of error. However, the GEP suggested a much shorter and more user-friendly model than ANN. Keywords: Self compacting concrete, mineral admixtures, artificial neural network, genetic programming, mixture optimization