Expansion tests for alkali-reactivity of aggregates – a comparison based on a kinetics approach

Abstract

The reactivity of aggregates regarding the alkali-silica reaction (ASR) has been measured by different properties affected by the reaction. One among those properties is the expansion measured by linear dilatation of mortar bars or concrete prisms. Several test-methods are used to determine the aggregate reactivity by this property. They differ in procedure for measuring, for accelerating the reaction, as well as in implementing conditions and criteria. Criteria were proposed usually by comparing with performance records of aggregates in field structures. The determination of reactivity, essentially a kinetic parameter, when measured by the same property, yields by different test-methods data that should be inter related, if converted by the effect of the varied parameters, to pre-fixed reference conditions. This paper tries to present such a kinetic inter relationship between standards NF P 18- 590, ASTM C 1260, and ASTM C 129. Globally, these standards cover a temperature range from 38 to 127 ºC, and different alkalinities. The relationship found shows consistency also for several aggregates, tested for NF P 18-590, ASTM C 1260, and ASTM C 227. To convert results of tests in standard conditions to a reference, common basis for comparison, at pre-fixed conditions, some assumptions were taken from the literature, to estimate the effect of alkalinity (given directly from the hydroxyl content or estimated from the alkalis content in the cement). Aspects related to the different experimental conditions on the test-methods and their effects are discussed. The methodology used in this work was based on the criteria comparison for aggregates with reactivity close to the critical one, for which some standards suggest a nearly linear expansion versus time. The proposed conclusion is that both the standards and the aggregate results, at the given conditions, are not inconsistent under the kinetic point of view. Some suggestions are made for improvements of accuracy of the relationship obtained.