Concrete Engineer, Laboratoire des chaussées, Ministry of Transportation of Québec, Sainte-Foy, QC
Research Assistant, Université Laval, Sainte-Foy, QC
Professor, Université Laval, Sainte-Foy, QC
Project Engineer, Direction Ingénierie, Hydro Québec, Montréal, QC
(Received 4 June 2001; accepted 5 April 2002)
The knowledge of the active- or soluble-alkali content of concrete is useful in the diagnosis and prognosis of alkali-aggregate reactivity (AAR). A method often used for determining this content is hot-water extraction from ground concrete samples. This method was applied to 17 aggregates and 8 concretes incorporating aggregates presenting different degrees of alkali-silica or alkali-carbonate reactivity. The following conclusions can be drawn: (1) a correction must be made to take account for the alkalies released by the aggregates in the test; (2) using cold water rather than hot water has no significant effect on the results; (3) grinding to <160 μm appears more appropriate than <80 μm (lower amount of alkalies released by the aggregates); (4) the soluble-alkali content progressively decreases as alkali-silica reaction (ASR) progresses, which indicates that a significant part of alkalies, progressively incorporated in the reaction products from ASR, is not leached in the test; (5) the repeatability from one series of tests to another and the reproducibility from one laboratory to another appear relatively poor; (6) the use of a control concrete with a known soluble-alkali content may greatly improve the repeatability and the reproducibility of the method.
Paper ID: CCA10489J