The use of alkaline activator in alkali-activated materials (AAMs) may pose risk of alkali-silica reaction (ASR), and the variations in the mixture design could have great influence on the performance of AAMs system. In this case, this paper investigated the effects of slag fineness (3000–8000 cm²/g) and water-to-binder (w/b) ratio (0.5–0.8) on ASR behavior of alkali-activated slag (AAS) mortars under accelerated mortar testing conditions as specified in ASTM C1260. The length change, mass gain, microstructure and formation of ASR products were examined to evaluate the degradation caused by ASR. It was found for the first time that slag fineness induces a “pessimum effect” in the ASR expansion of AAS mortars. On the other hand, there is a “pessimum effect” in the influence of w/b ratio on ASR expansion in the early-stage (≤14d), and the induced expansion increased with an increase in w/b ratio in the late-stage (>14d). The mechanism governing the effect of slag fineness and w/b ratio is complicated and cannot be explained solely by the properties of ASR products. This work contributes to the understanding of ASR in AAMs system and could provide a basis for the mixture optimization of AAMs.

Alkali activated materials (AAMs) have received worldwide attention due to its lower embodied energy and environmental impact than that of traditional cementitious materials. However, the activators with high alkalinity may raise the risk of alkali silica reaction (ASR) induced deterioration when reactive aggregates are used, which thereby limits the commercial use of AAMs. Not speaking the ASR induced long-term expansion, the early-age reaction of AAMs prepared with reactive aggregates is largely unknown. In this paper, isothermal calorimetry, thermogravimetry (TG) and mercury intrusion porosimetry (MIP) were adopted to study the heat evolution, mineralogical changes and pore structures of early-age ordinary Portland cement (OPC) mortar and water-glass activated slag/fly ash mortars. In each system, emphasis were made to understand the differences between mixtures prepared with standard inert quartz sands and reactive fine aggregates. The results show that the mortars prepared with reactive aggregates generated more heat in the wetting and dissolution stage. Particularly, the water-glass activated slag mortar presented the highest heat flow peak. Meanwhile, the results of TG illustrate that higher amount of reaction products were formed in water-glass activated mortars prepared with reactive aggregates than that with inert quartz sands. These findings suggest that the reactive aggregates are evidently involved in the early-age alkaline reaction of AAMs system.