Abstract

The accelerated mortar bar test (AMBT) is used to evaluate the potential expansion of the aggregate for the alkali-silica reaction. In this test, the aggregate should meet a certain distribution of particle size. In Uruguay it is usual that for concrete the fine aggregate is constituted by a mixture of sands, one of coarse size and another of fine size. Therefore, coarse sand must be crushed for the test and the fine sand is weighted accumulated in the smallest sieve in which there is enough material according to its particle size distribution. The aim of the work is to evaluate the influence of the particle size distribution of natural coarse sands, without crushing, in the result of the AMBT test. Mortars with different grading composition and single particle size were elaborated, and the differences obtained in the expansion, in the formation of efflorescence on the surface of the bars and in the presence of the reaction gel in the solution where the bars were immersed were analyzed. A greater expansion was observed in the samples with a continuous size distribution, compared with a single particle size. In addition, in the bars elaborated with the largest particle size, a greater amount of gel on its surface and in the solution was observed. The relevance of the work was to improve the reliability of the result in the AMBT test to evaluate the reactivity of the natural coarse sand. Natural sands of siliceous quartzfeldespathic origin were evaluated, without pessimum effect.