Statistical mixture design of ternary blends for controlling ASR

Abstract

Supplementary cementitious materials (SCMs) are long proven to control alkali–silica reaction (ASR) in concrete. Some SCMs such as slag and high-calcium fly ash are less effective than others and need high replacement levels, which may not be practical for various construction purposes. Combining two SCMs with Portland cement (i.e. ternary blend) offers design flexibility for the desired performance. This study investigates the use of statistical mix design in formulating ternary blends for ASR. Thirty-one mixtures were prepared based on statistical mixture design and tested using the accelerated mortar bar method. Five groups of ternary blends were investigated: Portland cement/metakaolin/brick powder, Portland cement/metakaolin/high-calcium fly ash, Portland cement/metakaolin/low-calcium fly ash, Portland cement/high-calcium fly ash/low-calcium fly ash and Portland cement/high-calcium fly ash/slag. The performances of the ternary blends were analysed using statistical models. The results showed that SCMs can be effectively combined in ternary blends to control ASR and the statistical mixture design can be a beneficial tool in formulating ternary blends for ASR performance. It was also found that 0·20% expansion at 28 d in an accelerated mortar bar test can be used as a second threshold for mixes containing SCM.