SECONDARY ETTRINGITE FORMATION AND DISTRESS IN A GROUT

摘要

Delayed ettringite formation (DEF) generally refers to an internal sulfate attack in portland cement concrete or similar materials in which the early normal formation of ettringite is postponed or interrupted, due to exposure to elevated temperature (greater than 70°C or 158℉) achieved either intentionally via heat-curing in precast products or inadvertently due to hydration, such as in mass concrete. If so affected, ettringite forms later after the concrete hardens and exposed to moisture. This delayed formation of ettringite generates substantial distress and results in expansion and subsequent cracking in concrete. Mitigation of DEF-related distress has to be considered in placing mass concrete and in production of precast concrete. Numerous publications are available on the topic. Two hypotheses have been proposed to explain the expansion caused by DEF: the uniform paste expansion theory and the crack expansion hypothesis. In the uniform paste expansion theory, the driving force is considered to be from the formation of submicroscopic ettringite within the bulk paste that causes expansion of the cement paste as a whole. The expansion creates peripheral gaps around aggregate particles, and the thickness of the gaps are proportional to the aggregate size. In the second hypothesis, the main driving force is considered to be local expansion resulted from the formation of ettringite in cracks in paste or rim cracks around aggregates where pressure comes either from an osmotic effect or from crystal growth. It is now generally believed that the paste expansion results from the growth of ettringite crystals in very small pores (～ 100 nm) of the cement paste (i.e. the uniform paste expansion theory) or both mechanisms may play a role (Taylor, 1997). Typical petrographic evidence of DEF includes: 1) frequent, partial or full peripheral gaps around aggregate lined with ettringite; 2) abundant cracks or microcracks within the paste that are lined with ettringite; 3) replacement of significant amounts of silicate paste by ettringite; and 4) presence of two-tone C-S-H inner products and variation in sulfate contents between the products. Some of these features, particularly the chemical composition, are not necessarily discernible using optical microscopy; thus, analysis using scanning electron microscopy equipped with energy dispersive X-ray spectrometry (SEM-EDS) is often needed. Exposure of concrete at the early age to elevated temperature has widely been accepted as the first prerequisite for DEF-related distress. Alleged DEF-related distress without heat exposure in some case studies had been reported, but was later found to be "unsubstantiated, inconclusive, or flawed" (Taylor, 1997). In our experience, the prerequisite of curing at elevated temperatures applies to concrete mixture with ordinary sulfate contents. For cementitious mixtures containing excessive sulfate, the heat curing for delayed formation of deleterious ettringite is not necessarily a required condition. This paper reports a case study that involves secondary ettringite-caused distress in an anchoring grout that was not subjected to elevated temperatures. A commercial, proprietary, pre-packaged cementitious product was used to anchor hundreds of aluminum handrails and posts at a Midwest train station. The grout is mainly composed of portland cement, calcium sulfoaluminate cement, and siliceous sand. The grout and the anchoring system exhibited apparent distress that caused displacement of the posts and cracking of surrounding concrete within five years from original installation. The product is intended for both exterior and interior use, and is expected to expand during setting rather than after setting, as the manufacturer claims. Our investigation, including field inspection, petrographic studies, X-ray diffraction analysis, and chemical composition analysis revealed that significant amounts of ettringite formed after the grout had set. Affected grout has altered substantially in physical, chemical and mechanical properties and eventually lost serviceability. Characteristics of the distress are similar to the heat-induced DEF, although the mechanism of ettringite formation is most likely very different. High mix water content in the grout, higher than manufacturer's recommendations, was suspected to have compromised the durability of the grout and contributed to the formation of excessive ettringite. Additional petrographic studies and mortar bar expansion tests were conducted on laboratory samples prepared using the commercial product to examine the hypothesis. The grout cylinders prepared with high water content (1.2 times of manufacturer's recommended) cracked in less than 40 days when saturated and exposed to the Chicago winter weather whereas the cylinders with the manufacturer's recommended water content and lower (0.8 time of recommended) stayed intact. Mortar bars did not expand but shrank up to 0.03 percent within the 3 months monitored when stored at an environment controlled room at 75℉ temperature and 75% relative humidity.