RILEM TC 258-AAA Round Robin Test: Alkali release from aggregates. Critical review of the test method AAR-8

The aim of this work is to introduce the method and modifications proposed by several laboratories after carrying out the Round Robin Test based on the AAR-8 method “Determination of alkalis releasable by aggregates in concrete”, which started in July 2015 [1]. This method was based on a literature survey on the subject, on a method first developed at Univ. Laval (Quebec City, Canada) and on other works developed at Institute Eduardo Torroja IETcc-CSIC (Spain), LNEC (Portugal) and LCPC (France). Five aggregates with different mineralogy and origin (phonolite, cataclasite, granodiorite, granite and a sea dredge basalt) were immersed in specific volumes of NaOH or KOH solutions with excess calcium hydroxide maintained at 38 °C and 60 ºC. At 2, 6, 13, 26 and 52 weeks, samples of the immersion test solution were removed by pipetting and chemically analysed for potassium (NaOH test solution) or sodium (KOH test solution). The amount of alkalis released were expressed in grams of Na2O, K2O, and Na2Oe per kilogram of aggregates. In this article, the results obtained of alkalis releasable by the tested aggregates are presented and, based in the obtained data, several changes are proposed in the test method in order to improve it. The main subjects of discussion were to avoid the agglomeration of fine aggregates and the alkali precipitation by reaction especially with fine aggregates, to promote the stabilization of the solution, to reduce the errors related to the method and mainly, to reduce the test time to 26 weeks instead of to 52. In addition, the test temperature is set at 38 °C due to no significant differences in the results obtained at 60 °C. The obtained results show that most of the aggregates tested release higher quantity of alkalis at 60 ºC than at 38 °C, except for cataclasite and granodiorite which present similar results at both temperatures. Also, it was observed a higher release of Na+ than K+, except for granodiorite and granite. Moreover, after 52 weeks of testing, the alkali release from aggregates has still not reached a plateau. The obtained results show a high dispersion between laboratories, which was attributed to the different amount of excess of calcium hydroxide used, the variation in the acidification of alkaline solutions before the chemical analysis, and the different analytical methods used by different laboratories as well as the calibrating method.