Microstructure and properties of fly ash/cement-based pastes activated with MgO and CaO under hydrothermal conditions

Abstract

Lime-activated fly ash cementitious materials are widely used as masonry cement and autoclaved building materials. This study investigated the effects of MgO on the microstructure and properties of the fly ash/cement-based pastes under hydrothermal conditions. The activator (i.e., CaO + MgO) was fixed at 17 wt%, and CaO was replaced with MgO at 0–30 wt%. The properties assessed were compressive strength, linear expansion, fluidity, porosity and pore size distribution. The microstructure was investigated by means of XRD, TG-DSC and SEM. Experimental results show that the compressive strength depended on the autoclave temperature, time and MgO content. Overall, as the MgO/(MgO + CaO) ratio was less than about 20, 15 and 12% at 140, 180 and 210 °C, respectively, MgO contributed to improve the compressive strength. In addition, appropriately extending the autoclave time or raising the autoclave temperature was in favor of strength development. The lower MgO content had a little influence on the expansion, but once the MgO/(MgO + CaO) ratio exceeded 20%, the expansion remarkably increased. Microstructural analyses demonstrated that MgO promoted the formation and crystallization of tobermorite likely through a solid phase or liquid-solid reaction mechanism. Besides, the micro-morphology of tobermorite tended to be slender in the presence of MgO.

Introduction

Fly ash (FA) is the main solid waste discharged from coal-fired power plants and contains active silica and alumina [[1], [2], [3], [4], [5]], it has good cementitious activity under alkaline condition and can be used as supplementary cementitious material in cement and concrete. Due to low cost and wide distribution in nature, CaO is usually used as an alkaline activator, then it is necessary to consume a large amount of high-grade limestone (CaO > 48%).

Unfortunately, the cement industry has occupied plenty of high-grade limestone. It is reported that 1.2 ton limestone is required for every 1 ton of clinker, therefore, it takes 2–3 billion tons of high-grade limestone a year to produce cement in China. Low-grade limestone is often accompanied by MgCO₃, which is decomposed into periclase (MgO) during the clinker burning process. The periclase will slowly hydrate into magnesium hydroxide after the cement paste is hardened. This process produces a molar solid volume expansion of 117%, which causes expansive stresses. As the expansive stresses exceed the tensile strength of cement paste, cracks occur, resulting in higher expansion [6]. For this reason, the MgO content is strictly limited to 5% in the cement industry [7]. This limits the utilization of magnesium-rich resources, which are generally be discarded. Therefore, it is of significance to study MgO as an alkaline activator instead of CaO.

At present, most applications of the CaO–MgO–SiO₂–H₂O cementitious system focus on the preparation of low-carbon cement, CO₂ sealant, and the expansion characteristics of MgO to achieve compensating shrinkage of cement and concrete at ambient temperature [[8], [9], [10], [11], [12]]. It was reported that reasonable addition of MgO could play the role of alkaline activator to promote the generation of hydration products [13,14]. On the other hand, MgO particles filled the pores to refine the pore structure of the cementitious materials. At the same time, Mg(OH)₂ generated by the hydration of MgO would cause volumetric expansion and internal defects. Yi et al. [15,16] investigated the CaO/MgO-slag pastes and found that the activated effects of MgO were better than those of CaO, the compressive strength measured at 28 d was 4 times that of cement pastes at the optimal MgO content of 5%–20%. The hydration products were mainly C–S–H gel, katoite and hydrotalcite-like phases. Ben Haha et al. [17] believed that the formation of hydrotalcite-like particles promoted the volume of hydration products, thereby reducing porosity and improving compressive strength. Gu and Jin et al. [18,19] found MgO was a potential activator for ground-granulated blast-furnace slag (GGBS), and the combined activator (0.5% CaO + 9.5% MgO) could enhance the early strength of the MgO/CaO-GGBS pastes. Besides, MgO was beneficial to the long-term strength development. The above studies indicate the replacement of CaO by MgO to prepare new cementitious materials can be feasible.

However, at present, there is still controversy over the type of hydration products and their formation processes in the presence of MgO. Vandeperre et al. [20] did not find the formation of hydrotalcite-like phases in the cement-fly ash-MgO system, so MgO played different roles in various cementitious system, which may be related to the surface solubility of slag or fly ash. Choi et al. [21] pointed out that 5% MgO slightly improved the compressive strength and durability of the cement-fly ash concrete. The total porosity of MgO-containing concrete was reduced resulting in a relatively dense microstructure, but the introduction of MgO had no influence on the fluidity of the fresh concrete.

In addition, the curing temperature is also a factor. Under autoclave conditions, C–S–H tends to crystallize, and more Mg(OH)₂ is likely to be produced, leading to the volume expansion. Ali and Mullick [6] found that fly ash could inhibit the expansion of the high magnesia cement and improve its mechanical strength at high temperature. Qian et al. [22] compared the expansion inhibition effects of several siliceous materials on the MgO-cement system, and demonstrated that the effect order was slag < fly ash < quartz sand. The formation and quantity of crystalline C–S–H (tobermorite and xonotlite) had an impact on the inhibition of volume expansion at high temperature. Nakagawa et al. [23] believed that MgO promoted the formation of tobermorite in the cement-silica fume-slag pastes at 175 °C, and accompanied by expansion. Further, xonotlite was produced as the temperature increased to 250 °C, moreover, the expansion was significantly increased. However, the effects of co-existence of MgO and CaO on the microstructure, mechanical strength and volume deformation of the CaO–MgO–SiO₂–H₂O cementitious system under hydrothermal conditions are still lack of comprehensive studies. In addition, the effects of MgO on the rheological properties of composite cementitious system have been rarely reported.

The objective of this work is to investigate the effects of MgO on the microstructure and properties of the fly ash/cement-based pastes under hydrothermal conditions. The hydration products were evaluated by XRD and TG-DSC, and the morphology of hydration products was observed by SEM. The porosity and pore size distribution of the pastes were measured by the MIP method. The variations in compressive strength and linear expansion were interpreted by changes in the pore structure and hydration products. Finally, the formation mechanism of the hydrothermal products in the pastes was discussed. Our work has theoretical and practical significance for applications of magnesium-containing carbonates and magnesium-rich resources in the autoclaved building materials, thermal insulation boards, oil well cement and fiber-reinforced cement-based products.