1. Introduction
In recent years, there has been an important increase in the use of self-compacting concrete (SCC) [1-6]. Since its emergence, SCC is widely used all over the world. SCC was developed in Japan to improve the uniformity and reliability of concrete [7]; it doesn’t require any consolidation work at site. The characterization and for- mulation of this material have been the subject of numerous investigations [8-11].
Using SCC in structures would result in both technical and economical advantages. One of the most important differences between SCC and conventional concrete is the incorporation of a mineral admixture. This concrete is characterized by a high amount of fines, an amount of water, a relatively low use of super plasticizers, a high deformability and good uniformity in such a way that it can flow under its own weight to completely fill the formwork and passes through the congested reinforcement without any mechanical vibration. Many studies show the advantage of mineral admixture usage in SCC; and it enables to improve the workability with a reduction of cement content [12-14]. The mineral admixtures enable to improve particle packing, to decrease the permeability and to increase the durability of concrete [15]. The waste materials such as limestone powder, fly ash and granulated blast furnace slag are generally used as mineral admixtures in SCC [16-19] These add a positive impact on the timeliness and quality of concrete [7], at the same time the environmental pollution will be reduced [20].
The aim of the present work is to highlight the influence of local constituents in the composition of a SCC from the viewpoint of fresh and hardened state behavior and therefore to develop optimized formulations with good rheological and mechanical properties.
This paper deals with the investigation of the effect of LP, BFS and SF as mineral admixtures on the fresh and hardened properties of SCC. Therefore, the saturation point with the cone Marsh, the slump flow, the compressive strength, the ultrasonic pulse velocity (UPV) and the dynamic elastic modulus tests were conducted to achieve this objective and determine the appropriateness of using these different material admixtures in SCC.
The experimental work began with the characterization of various local materials from Algeria.
2. Experimental Program
2.1. Materials
The materials used in this study were locally sourced and they satisfied the requirements of Algerian Standards.
2.1.1. Cements
Portland cement (CPA CEM I 42.5) is supplied by the cement—Ain Kebira—Algeria, according to EN 197/1 (European Committee for Standardization—2000) [10]. Its mineralogical composition is C3S = 61.3%, C2S = 15.9%, C3A = 8% and C4AF = 9.6%.
2.1.2. Aggregates
The coarse aggregates are from a deposit of limestone (Kef-Erendira region, Eastern of Algeria). Its Los Angeles coefficient and absolute density are respectively 20% to 28%, and the value of absolute density is equal to 2.7. Natural sand, (fine aggregates) was procured from a nearby river source. It has a sand equivalent, fineness module and absolute density equal to 78%, 2.0 and 2.60 respectively.
2.1.3. Minerals Addition
The addition included minerals are Silica fume with blain specific surface of 200,000 cm2/g, local limestone fillers (type Alcal15), blast furnace slag (Blaine specific surface of 4200 cm2/g).
2.1.4. Super Plasticizers
Super plasticizer by trade name MEDAFLOW 30 was used as high water reducing agent, third generation, to achieve the required workability.
2.2. Principles of Mix Proportioning
The portioning of the mix is extremely important in developing an effective SCC. The effect of mixture proportion, in terms of ratios between cement to solids and coarse and fine aggregates has been the subject of many investigations [21].
The incorporation of fine mineral remains empirical. There is no universally agreement on the effect of these factors due to the complexity of combined action.
Assuming that fresh concrete behaves as a material in two phases namely a viscous phase, consisting of the paste (cement + fines + water) and a granular phase containing all the aggregates. Preliminary concrete mixture (Table 1), was adopted as a control mix test; and the two phases are analyzed.
2.3. Specimen Preparation
The concrete was mixed using a mixer with a capacity of 100 l. The mixture sequence was as follows: all the solid components (natural aggregates, cement, silica fume, blast furnace slag and calcareous filler) were mixed for 30 s. It was followed by the introduction of the effective water and one third of the super plasticizer. After 90 s of mixing, the remaining additives (the rest of the super plasticizer and the viscosity agent) were added and mixed with the other components for 210 s. Then the slump flow tests were carried out. Depending on the obtained test results, additives and/or water were adjusted. Specimens were cast in various moulds of different shapes according to the test requirements. Due to the flowing ability of fresh SCC, the materials were successfully poured directly into the moulds without vibration.
Table 1. Preliminary concrete mixture.
2.4. Characterization of Fresh Concrete
2.4.1. Optimization of the Paste
Optimization of the bonding paste is obtained using the Marsh cone test. The study focuses on the variation of flow time of different grout depending on the dosage of super plasticizers. Four grout mixtures were investigated. First mix don’t contain any added mineral, the second one contains 10% of silica fume, the third mix contains 30% of limestone filler and the last mix contains 30% of blast furnace slag.
Figure 1 shows the variation of flow time for different mixes. It decreases with the increasing of super plasticizer dosage until reaching a certain point called the saturation point. After that, it remains nearly constant or with a slightly increase.
The flow time of the mix without mineral addition is 0.9% by weight of dry cement (Figure 1(a)). It is 1% for the mix with silica fume (Figure 1(b)), while it is 1.1% in the cases of the two others mixes (Figures 1(c) and (d)).
2.4.2. Optimization of the Granular Skeleton
The development of an effective SCC involves either modifying the cement paste, or tuning the aggregates, or both of them [20].
The interlocking of coarse aggregates is integral to the strength of the concrete [22].
With coarse aggregates, changing inter-particle spacing most practically changes the flow ability of concrete. Coarse aggregates tend to settle with the introduction of super plasticizer, which causes segregation.
Two proportions were studied in fresh state of SCC. The first mix contains 67% of gravel (3/8) and 33% of coarse aggregates (large gravel) (8/15) while the second one contains 33% of gravel (3/8) and 67% of coarse aggregates (8/15).
The mixtures of three sets of concrete for three different sand/paste ratios were manufactured with two mixes for each series. Only the quantities of fine aggregates and coarse aggregate were varied, while the other constituents were kept constant for the entire investigation. The water to binder ratio was also kept constant. The details of different series are given in Table 2.
Table 2 and Figure 2 show the effect of aggregate content on the large diameter of slump for water to binder ratio equal to 0.4.