PSI - Issue 37

João Custódio et al. / Procedia Structural Integrity 37 (2022) 590–597

593

4

João Custódio et al. / Structural Integrity Procedia 00 (2019) 000 – 000

4. Results and discussion 4.1. Aggregate potential alkali-reactivity – Expansion tests The results obtained in the expansion tests carried out to determine the aggregates’ potential alkali-reactivity are presented in Fig. 1. It is important to highlight that according to RILEM AAR-0 (RILEM, 2016b), if the petrographic analysis made to the aggregate material, following RILEM AAR-1.1 (RILEM, 2016c), does not detect known alkali-reactive rock types and/or mineral species in significant amounts, the aggregate is classified as Class I (very unlikely to be alkali reactive), and no further actions are needed; being allowed that the aggregate is used in the field concrete. On the other hand, if an aggregate is classified as Class II (alkali-reactivity uncertain – an aggregate material that cannot be unequivocally classified as either Class I or Class III) or Class III (very likely to be alkali-reactive – an aggregate material that contains alkali-reactive rock types and/or mineral species in sufficient amount known to cause damage in concrete), it is necessary to perform (i) the accelerated mortar bar test, AMBT (RILEM AAR-2), in the case of aggregates designated as Class IIS or IIIS, or (ii) both the AMBT (RILEM AAR-2) and the concrete microbar test (RILEM AAR-5), in the case of aggregates designated as Class IISC, IIICS, IIC or IIIC. If in the test AMBT (RILEM AAR-2) the aggregate is considered to be reactive (expansion greater than 0.20 % at 16 days) or potentially reactive (expansion in range 0.10 – 0.20 % at 16 days in the absence of previous local use experience) than the accelerated concrete prism test (ACPT) (RILEM AAR-4.1) or the concrete prism test (CPT) (RILEM AAR-3.1) must be performed. The criteria for interpreting the results from RILEM AAR-3.1 and AAR-4.1 test methods have not yet been finally agreed (RILEM, 2016b). However, on the basis of trials carried out by RILEM on aggregate combinations of known field performance from various parts of the world, it appears that results in the AAR-3.1 test (usually at 12 months of testing) of less than 0.05 % are likely to indicate non-expansive materials, whilst results exceeding 0.10 % indicate expansive materials; while in the case of the AAR-4.1 test, it seems that a maximum expansion in the test of 0.03 % at 15 weeks indicates a non-reactive aggregate combination (RILEM, 2016b). In the case of aggregate combinations producing AAR-3.1 results in the range 0.05 % to 0.10 %, after 12 months, or AAR-4.1 results greater than 0.03 %, at 15 weeks, and in the absence of previous local experience on their ASR reactivity, they should be regarded as being potentially alkali-reactive, and precautionary measures should be used to minimise the risk of ASR damage to any concrete in which they are used (RILEM, 2016b). There is some evidence that a lower AAR-3.1 criterion at 12 months (perhaps 0.04 or even 0.03 %) might be applicable for some slowly reactive aggregates (RILEM, 2016b). The most recent maximum expansion limits, suggested by RILEM, for concrete used in very large, long-service structures, like dams, are (RILEM, 2016a): AAR-3.1 – 0.03 % at 1 year, and/or 0.04 % at 2 years; AAR-4.1 – 0.02 % at 15 weeks, and/or 0.03 % at 20 weeks or longer ( e.g. , 0.03 % limit at 1 year). In Portugal, the new LNEC Specification E 461 (LNEC, 2021b) states that aggregates, other than granites and vulcanites, may be accepted as non-reactive solely on the basis of petrography, when alkali-reactive forms of silica are not observed in the aggregate during the petrographic analysis (aggregate is then classified as belonging to Class I – very unlikely to be alkali-reactive), but that decision has a certain risk associated; therefore, in some situations, for instance when the aggregate is to be used in the concrete for structures like dams and other very long service life structures, it is advisable to perform the expansion tests as well. However, if potentially alkali-reactive forms of silica are indeed observed in aggregates, other than granites and vulcanites, during the petrographic analysis, then it is mandatory to perform at least the accelerated mortar bar test (AMBT – ASTM C1260 or RILEM AAR-2). In the case of granites and vulcanites, where the AMBT is not recommended, it is always necessary to assess the extent of the alkali-reactivity and appropriate levels of prevention, hence being mandatory to perform the accelerated concrete prim test (ACPT – RILEM AAR-4.1) or the concrete prim test (CPT – RILEM AAR-3.1). If the aggregate or aggregate combination assessed with the ACPT results in an expansion, of the concrete prisms, lower or equal to 0.02 % at 13 weeks or 0.03 % at 20 weeks, they are classified as Class II (Class II – potentially alkali-reactive or alkali-reactivity uncertain); however, if the expansion is higher, then the CPT must also be performed. If the aggregate or aggregate combination assessed with the CPT results in an expansion, of the concrete prisms, lower than or equal to 0.03 % at 12 months or 0.04 % at 24 months, they are judged as Class II (potentially alkali-reactive aggregate (or aggregate combination) or aggregate (or aggregate combination) of alkali-reactivity uncertain); conversely, if the expansion is