Can certain alkali minerals explain the slow reactivity of granitic aggregates in dams?

Abstract

Swelling processes due to alkali-silica reactions are a major limitation to durability of concrete dams and hydraulic structures. These reactions occur in high humidity environments between the aggregates and the alkaline cement interstitial fluids, resulting in the formation of expansive alkali-silica gels. The speed, or even the occurrence, of these reactions is quite difficult to predict because it is not related simply to the mineralogical nature of the aggregates. A mineral can even prove to be stable or unstable in different rocks and the relative reactivity of various minerals is also different from aggregate to aggregate. So, there is controversy about which are the factors (chemical composition, structural type, micro-deformation, …) that effectively control the alkali reactivity observed in some aggregates. Besides, there is no consensus about what kind of alkaline minerals are more likely to release alkalis to the interstitial cement solution, and also on the factors that accelerate that solubilisation. This paper aims to contribute to the ongoing discussion of this topic, and so different granitic aggregates, with sound and altered factions, were subjected to alkaline solubility tests in order to measure the alkalis released and to identify the mineral phases that contribute to alkali media raise. The results obtained confirm that granitic aggregates release alkalis in alkaline environment, being this release dependent on the degree of alteration of these aggregates. Natural altered aggregates have less capacity to release alkalis than the non-altered aggregates. In the tested granite samples we observe a higher release of potassium, which is attributed to the higher natural alteration of plagioclase in comparison to K-feldspars.