6. Conclusions
The results obtained from the four test methods including expansion, nonlinear acoustics, microwave measurements, and quantitative petrographic analysis using the damage rating index (DRI) provide a better picture of how expansion and damage are linked to the gel formation during the AMBT test. The main findings of this study are as follows: Among the dielectric properties of the samples, relative permittivity shows higher sensitivity to ASR gel formation, and the samples cast with the potentially reactive aggregates show higher relative permittivity than those prepared with innocuous aggregate. DRI values increase over the time of exposure to the AMBT condition, and increase consistently with the expansion data for samples cast with potentially reactive aggregates, while this is not the case for the samples prepared with innocuous aggregate. Not only does the cumulative average nonlinearity parameter strongly correlate with the expansion of each sample type, but also this correlation is stronger for samples cast with potentially reactive aggregates than those cast with innocuous aggregate. The slope of the linear relationship between the cumulative average nonlinearity parameter and expansion is approximately the same for samples cast with the potentially reactive aggregates, while it is distinct for those prepared with the innocuous aggregate. The correlation between the cumulative average nonlinearity parameter and DRI is stronger for samples prepared with potentially reactive aggregates than those cast with the innocuous aggregate. The agreement of the relative permittivity and DRI of sample types may originate from the larger weights that the DRI assigns to the defects containing ASR gel, and the sensitivity of the relative permittivity to the gel presence.