ALKALI-SILICA REACTION AND SURFACE DISTRESS

Alkali-silica reaction (ASR) is a distress mechanism associated with certain aggregates in portland cement concrete. ASR occurs when microcrystalline, glassy, or physically strained silica (all chemically SiO2) reacts with alkaline pore solution in portland cement concrete and generates an expansive gel when the gel absorbs water. WJE frequently encounters ASR issues in our troubleshooting using petrographic studies.

Small amounts of potentially ASR-reactive particles are common in a lot of aggregates and may or may not be destructive depending on reactivity, quantity, and size of these materials, alkali contents, and moisture exposure, among many other factors. Similar amounts of deleterious materials tend to more frequently cause unacceptable aesthetic, performance, and even safety concerns when they occur at the concrete surface or in the near-surface region compared to the bulk of the concrete. For example, surface blisters and uneven surfaces of epoxy coatings, vinyl tile, or linoleum flooring may lead to performance and aesthetic issues. Recently, WJE has worked on a few projects that involved ASR-caused blisters and allegedly ASR-caused blisters of industrial, medical, and airport hangar floor finishes. These case studies illustrate the usefulness of petrography at all stages of an investigation, from materials screening to identifying the root cause of the distress.

Case Studies

Two flooring blistering cases occurred in Colorado, 120 miles apart. Alkali-silica reaction of rhyolite, a volcanic rock with high amorphous and microcrystalline silica (an extrusive rock equivalent to the plutonic granite) that is frequently present in the local gravel used as coarse aggregate, was identified to be the culprit in both cases. Another case occurred in San Antonio, Texas, and chert in the fine aggregate was identified as the ASR-reactive component responsible for the blistering. In these projects, blistering was observed in the floor coatings from a few months to approximately two ears after installation of the flooring systems. In two of the case studies, the clients claimed that the blistering problem had recurred repeatedly after repair or replacement of the flooring system. Our studies revealed that the root cause of the blistering was ASR in the near-surface concrete and that replacing the flooring system would therefore be ineffective against continued near-surface ASR in the concrete. In the third case study, blistering occurred only in cut-out and subsequently repaired concrete areas where a coarse aggregate containing rhyolite had been used in the repair concrete. In all three cases, the content of reactive component was estimated to be less than 3 percent of the total aggregates. Generally, a reactive particle (rhyolite or chert) was observed below each blister that had undergone ASR; the size of a surface blister generally depended on the size and depth of the offending aggregate particle. Microcracks filled with ASR gel emanated from the affected particles and extended to the concrete surface. ASR gel was frequently observed at the blister area between the base coat of the flooring system and the concrete surface. The near-surface reactive particle occasionally exhibited material loss along the internal cracks due to consumption caused by reactions. Microcracks in the concrete were usually more abundant below and near the blistered area than in the coated surface away from the blister.

In the two rhyolite-involved cases, ASR gel and associated microcracking were also observed in the body of the concrete. Exposed rhyolite particles and paste in their vicinities frequently appeared wet or glossy due to fresh production of ASR gel. In the chert-involved case and many other reported cases, ASR occurred only in the near-surface region. Bleeding in fresh concrete is generally considered to elevate concentrations of alkalis in the surface region and promote near-surface ASR. Previous WJE investigations on different cases (e.g., Cong et al., 2004; Lawrence and Cong, 2003) analyzed the depth profile of the alkali distribution and revealed greater water-soluble alkali concentrations toward the top surface. Low permeable (vapor-tight) floor covering—present in our case studies—and materials such as surface sealers, cleaning solutions, or other applied materials may also increase moisture content and alkali levels in the top region and exacerbate ASR. Chemical analysis of blister fluids collected below blisters may reveal alkalinity of the fluids and provide useful information regarding the species and source of elevated alkalis. The fluids frequently have a pH up to 14 (highly alkaline). Many previous studies have also emphasized the role of fine aggregate in near-surface ASR and other consultants concluded that coarse aggregate was generally less involved in distress caused by near-surface ASR. Our case studies did not support this claim.

Conclusions

Caution has to be taken to positively attribute surface distress to ASR. Distinction needs to be made between observations and interpretations, between the presence of potentially reactive aggregate and evidence of actual reaction, and—further—between minor non-harmful reaction and severe deleterious reaction or significant distress (cracks). Reactive aggregate, ASR gel, associated cracks, and spatial connection between the blister and reactive particle are required (although not necessarily sufficient) conditions/characteristics for ascribing surface distress to ASR. In a few other cases we studied or reviewed, blistering, debonding, or delamination was unrelated to ASR. However, consultants retained by the other parties misidentified the root cause of the distress and blamed surface ASR, based solely on the presence of potentially reactive aggregate and/or by mistaking surface hardeners or even adhesives for ASR gel.

A combination of field investigation, proper sampling and preparation, petrographic examinations, and chemical analysis is key to identifying ASR-caused surface distress. Petrographic studies (ASTM C856, Standard Practice for Petrographic Examination of Hardened Concrete) of the floor concrete is recommended before installing a flooring system to assess the potential for ASR-related distress, determine the need for or adequacy of subsequent surface preparation, and evaluate anticipated performance of various flooring components and potential failure locations. Testing in compliance with ASTM C457, Standard Test Method for Microscopical Determination of Parameters of the Air-Void System in Hardened Concrete, can be used to determine the volumetric content of reactive particles in hardened concrete. Petrographic examinations in accordance with ASTM C295, Standard Guide for Petrographic Examination of Aggregates for Concrete, can be conducted on aggregate samples to identify potentially alkali-silica reactive (and other deleterious) constituents, determine such constituents quantitatively, and recommend additional tests to assess aggregate constituents of poor performance. Information gathered through these assessments can be used to adjust mix components to reduce problems prior to construction or to address anticipated problems in existing construction.

Credit to: Hugh Hou, Rich Cechner, Terry McGovern, Kevin Michols, and Todd Nelson