To be or not to be... alkali-reactive. A challenge for the petrographic method

Abstract

The mechanisms involved in alkali-aggregate reaction and the characteristics of reactivity of some aggregates are not completely understood, although tens of thousands of mortar and concrete prisms have been tested in the laboratory and field exposure sites have been installed in a number of countries. Aggregates are the least well understood component in the alkali-aggregate reactions and, surely, the most difficult to assess. This happens not just because the materials used as aggregates are widely variable in origin, geological history, composition, texture and degree of alteration, but also because rocks that behave as innocuous in certain areas of the world appear to perform as reactive elsewhere. In consequence, the petrographic assessment of the aggregates is commonly not enough to classify an aggregate as innocuous or potentially reactive. In this paper questions are raised concerning slow reactive rock types, including suggestions about the possible forms of silica present and their identification.