Impact of AAR-reactive fillers on ASR-induced development

Abstract

The NetZero target has stimulated studies focused on reducing CO2 emissions in several industries. In the cement industry, one of the approaches is the use of supplementary cementitious materials and aggregate mineral fillers (AMFs) to partially replace cement. However, some of the rocks used to produce AMFs may be susceptible to alkali-aggregate reaction (AAR), a harmful distress mechanism affecting critical concrete infrastructure, and their impact in concrete durability remains unclear. Considering this context, this study aimed to contribute to comprehending the effect of AAR-reactive AMFs on AAR-induced damage in concrete. Initially, a thorough literature review was conducted to understand the current state of the art and identify knowledge gaps concerning the use of AAR-reactive AMFs, focusing on the roles of mineralogy, particle size distribution (PSD), replacement content, and the test methods used to assess AAR kinetics, ultimate expansion, and associated microscopic and mechanical deterioration. Based on the gaps found, an experimental program was

designed to assess the influence of alkali-silica reaction (ASR)-reactive AMFs on ASR- induced damage in concrete when used to replace cement in systems with reactive

and non-reactive aggregates. Three systems were evaluated: one with non-reactive aggregates (mixtures A), one with reactive coarse aggregates (mixtures B, Springhill), and one with reactive fine aggregates (mixtures C, Texas sand). Two reactive rocks were used to produce fillers, one moderately reactive mylonite and one highly reactive greywacke, with varying PSDs (i.e., <150 μm and <75 μm) and used at cement replacement rates of 10% and 20%. The concrete prism test (CPT) was adopted to monitor expansions and kinetics. To evaluate the deterioration process, the multilevel assessment was applied by means of microstructural (i.e., damage rating index – DRI) and mechanical (i.e., stiffness damage test – SDT) evaluations at 1, 3, and 6 months. For the systems with reactive aggregates, results indicated that ASR-reactive AMFs altered the kinetics of the systems in different ways, which was related to the competition for alkalis at early ages. The expansion rate reduced, tending to stabilization from 180 and 60 days for mixtures B and C, respectively. Mechanical degradation remained significant, as evidenced by increased stiffness damage index

values and reduced modulus of elasticity over time. In general, the introduction of ASR- reactive AMFs in a system with reactive coarse aggregates did not alter the ultimate

expansion and degradation when compared to a system with the same aggregate with

no fillers. For a system with reactive fine aggregates, the ultimate expansion was reduced but the degradation level was significant when compared to a system with no fillers. For the system with non-reactive aggregates, the filler reactivity, PSD, and percentage significantly affected expansion behavior, with finer PSD materials (<75 μm) promoting higher expansions. Despite some expansion, the values remained in general below the threshold set by standards. Moreover, negligible physical damage was observed. The findings indicate that ASR-reactive AMFs can be a viable alternative for blended cements and suggest that CPT combined with multilevel assessment is a reliable approach for evaluating ASR-reactive AMFs.