**Sung Choi and Kwang-Myong Lee \***

Department of Civil, Architectural, and Environmental Systems Engineering, Sungkyunkwan University, Jangan-Gu, Suwon 16419, Korea

**\*** Correspondence: leekm79@skku.edu; Tel.: +82-10-8207-7516

Received: 17 May 2019; Accepted: 26 June 2019; Published: 27 June 2019

**Abstract:** The performance of alkali-activated slag (AAS) paste using activators of strong alkali components is affected by the type, composition, and dosage of the alkaline activators. Promoting the reaction of ground granulated blast furnace slag (GGBFS) by alkaline activators can produce high-strength AAS concrete, but the workability might be drastically reduced. This study is aimed to experimentally investigate the heat release, workability, and setting time of AAS pastes and the compressive strength of AAS mortars considering the Na2O content and the ratio of Na2O to SiO2 (Ms) of binary alkaline activators blended with sodium hydroxide and sodium silicate. The test results indicated that the AAS mortars exhibited a high strength of 25 MPa at 24 h, even at ambient temperature, even though the pastes with an Na2O content of ≥6% and an Ms of ≥1.0 exhibited an abrupt decrease in flowability and rapid setting.

**Keywords:** alkali-activator; GGBFS; Na2O content; Ms (SiO2/Na2O); workability; setting time

### **1. Introduction**

Environmental imperatives such as the reduction of CO2 and conservation of natural resources are becoming issues worldwide. However, because the cement industry emits large amounts of CO2 and consumes much energy in the supply of raw materials and production processes, efforts are needed to reduce CO2 emission and energy consumption by replacing cement with supplementary cementitious materials (SCMs) such as ground granulated blast furnace slag (GGBFS) and fly ash (FA). Thus, in order to totally replace cement with GGBFS and FA attempts have been made to create zero-cement concrete by the alkali activation of GGBFS and FA [1–5].

Alkali-activated slag (AAS) concrete, which is produced by using an alkali activator on GGBFS, is a typical zero cement concrete. As GGBFS is calcium aluminosilicate vitreous, it has strong latent hydraulic properties. However, alkaline (pH = 12) activators are needed to stimulate GGBFS. After the GGBFS is stimulated, Si4<sup>+</sup>, Al3+, Ca2+, Mg2+, and Na<sup>+</sup> ions are eluted from GGBFS and the reaction of GGBFS proceeds. It should be noted that GGBFS activated in this way has very fast reaction rates due to chemical ionic reactions [6]. Alkaline activators largely impact the properties of AAS concrete and thus, their characteristics such as the alkali concentration, the dosage, and the water-to-solid ratio should be examined prior to the mixture design of AAS concrete [7,8]. In general, soluble alkali or alkali salts can be used as alkaline activators. When considering the strength development and economic efficiency of AAS concrete, the most commonly used activators are Na2SiO3 (sodium silicate) and NaOH (sodium hydroxide). A better understanding of the effects of such alkaline activators on the reaction mechanisms of alkali-activated slag could indicate ways to optimize the use of alkaline activators.

Recently, research has been conducted to improve the performance of AAS concrete by blending two types of alkaline activators [9]. When such a binary alkaline activator is used, the characteristics possessed by each alkaline activator can be complementarily utilized [10,11]. NaOH promoted initial reaction of AAS, and Na2SiO3 was effective in increasing the strength of AAS concrete. Chang [12] evaluated the setting characteristics of silica-activated pastes depending on the type of alkaline activator used and concluded that setting time significantly reduced as the amount of alkaline activator increased. Zuo et al. [13] showed that though the reaction degree increased as the amount of alkaline activator increased, the reaction degree no longer increased when the amount of alkaline activator was greater than 6 wt.% of cement. Krizan and Zivanovic [14] reported that high strength could be expected when the Ms (SiO2/Na2O) were in the range of 0.6–1.5. As described above, as the amount of alkaline activator increases, the reaction of GGBFS increases and the strength of AAS concrete also increases. Therefore, it is necessary to determine an appropriate range for the usage amount of the activator for the mix design of AAS concrete. In particular, the effects of Na2O and SiO2 concentrations in alkaline activators on the strength development of AAS concrete should be examined in order to effectively use the alkaline activators.

AAS concrete, which stimulates GGBFS using alkaline activators, rapidly increases in strength in the early ages and exhibits high strength of 40 MPa or more at 28 days [15,16]. However, because the initial reaction rate of GGBFS by alkaline activators is extremely high, several problems with AAS, such as the flowability being lost initially or rapid setting occurring, must be improved for it to be used in a practical setting [17–19]. To address this, the effects of the alkaline activator on the fresh properties of AAS concrete such as workability and setting should be examined, even though the AAS activation mechanism at early ages has not been clearly identified.

In this study, in order to identify the characteristics of AAS pastes, a binary alkaline activator that was a blend of NaOH and Na2SiO3 was used. Nine types of AAS paste mixture, varying by their Na2O content and the ratio of Na2O to SiO2 (Ms), were tested. First, the reactivity of the AAS was evaluated for 24 h through the calorimetric measurement of the AAS pastes, and the workability and setting time were measured. Furthermore, to quantitatively analyze the instant at which flowability is lost and setting occurs in the AAS pastes, the viscosity and ultrasonic pulse velocity of the pastes were measured. Lastly, the compressive strength of the AAS mortars was measured up to 24 h. Consequently, it was found that the workability and setting of AAS pastes and the initial strength development of the AAS mortars were significantly affected by Na2O content and Ms.

## **2. Materials and Method**
