PSI - Issue 37

João Custódio et al. / Procedia Structural Integrity 37 (2022) 644–651 João Custódio et al. / Structural Integrity Procedia 00 (2019) 000 – 000

645

2

1. Introduction Nowadays, the internal sulfate reaction, ISR (term hereinafter used to designate the phenomenon heat-induced internal sulphate attack, HISR, which is also referred in the literature as delayed ettringite formation, DEF) still constitutes one of the main causes of concrete degradation in massive concrete elements or structures. ISR a form of sulfate attack by which mature hardened concrete is damaged by internal expansion during exposure to cyclic wetting and drying in service and caused by the late formation of ettringite, as a result of a chemical reaction between sulphate ions and calcium aluminates present in the hardened cement paste. ISR is not likely to occur unless the concrete has been exposed to temperatures during curing of about 70 ºC or greater and less likely to occur in concrete made with pozzolan or slag cement (ACI, 2013). In ISR, as the name suggests, the source of sulphate is in the concrete and can be the cement, the supplementary materials ( e.g. , fly ash), the aggregate, or the chemical admixtures. The formation of the so-called secondary or delayed ettringite has an expansive nature and can cause the disruption of concrete (Taylor, 1997, Skalny et al., 2001, Taylor et al., 2001). If sufficient ISR occurs in the concrete, the induced pressures cause microcracking and then expansion of the surrounding concrete. The build-up of stresses that cause cracking can be local and at macroscopic level. Locally due to the differential expansion in constituents, paste and aggregates. At larger scale, caused by different expansion conditions in the same structural element, due to differential macroscopic swelling, or caused by expansive deformations in hyperstatic structures. At the macroscopic level, the concrete surface does not expand to the same extent as the interior, because the conditions required for the reactions are not totally fulfilled at the concrete surface, for example, the concrete surface is subject to leaching of the alkalis and to temperatures lower than those felt in the bulk of the concrete mass. This causes tensile stresses to arise in the surface and induce surface macro-cracks. The formation and orientation of both micro and macrocracks are affected by internal or external restraints that also reduce expansion. Generally, ISR generates a very significant drop in terms of tensile strength and modulus of elasticity, and an important decrease in the compressive strength of concrete (Custódio and Ribeiro, 2015a, Custódio and Ribeiro, 2015b, Custódio, 2017, Custódio and Ribeiro, 2019, Custódio, 2020). Hence, the structural integrity of large concrete structures can be severely jeopardized by ISR evolution, which ultimately can lead to their decommissioning and demolishing. Currently, in most practical situations, there is no effective way of stopping ISR, however, in some cases it can be slowed down through rehabilitation works. Therefore, ISR prevention is of upmost importance to prevent costly rehabilitation procedures or even the need to decommission/replace the affected structure. The most effective way to prevent the deleterious development of ISR is to control the maximum temperature attained during concrete cure. The control of temperature can be made (i) with an adequate mixture design, and (ii) by controlling temperature during concrete production, concrete transport, concrete pouring and after the concrete has been placed and compacted. The tools currently available to practitioners, for predicting the maximum temperature attained in a concrete element, require the previous knowledge of the heat of hydration of the cement to be used in the structure. Several methods can be used to determine the cement heat of hydration, although the most common has been the Lagavant method, i.e. by means of semi-adiabatic calorimetry, the isothermal conduction calorimetry method is now becoming a widely used method. Since the two test methods produce different values, it is important to establish a correlation between the results obtained with both and to determine their influence on the temperature estimate made with the currently available calculation tools. This paper aims at addressing these concerns by presenting the heat of hydration determinations made with the two methods to several cement types normally used in concrete structures, the correlation obtained between the two methods, the temperature estimates made with them, and their comparison with the actual temperatures recorded in a structure. 2. Laboratory testing campaign The laboratory testing campaign includes the determination of the heat of hydration of 25 cement samples, which correspond to 13 different cements, and some of them with samples from different batches; the determinations were made with two methods – semi-adiabatic calorimetry and isothermal conduction calorimetry. The laboratory campaign also included the determination of the potential ISR reactivity of the concrete further assessed in-situ (section §3). The global laboratory testing campaign is summarized in Table 1.