Early-age GGBS concrete hydration temperature development and modelling

Abstract

Early-age hydration temperature development in concrete plays a crucial role in determining its structural performance and long-term durability. Excessive temperature rise and thermal gradients can induce stresses, leading to early-age cracking, particularly in mass concrete and large-span structures. The incorporation of Ground Granulated Blast Furnace Slag (GGBS) as a supplementary cementitious material helps mitigate these risks due to its lower heat of hydration compared to CEM I. However, knowledge gaps remain regarding the effects of variable ambient temperatures, the influence of coarse aggregates, and the applicability of existing hydration models developed for CEM I-only concrete. This study addresses these gaps through experimental investigations and numerical modelling. The experimental program involved semi-adiabatic and isothermal calorimetry tests. Semi-adiabatic calorimetry tests on concrete specimens with varying GGBS replacement levels (0%–50%) assessed the impact of GGBS on hydration temperature development under uncontrolled conditions. Results demonstrated that increasing GGBS content reduced peak hydration temperatures and prolonged the induction period, confirming its thermal mitigation effect. Isothermal calorimetry tests were conducted on micro-concrete and equivalent mortar specimens at curing temperatures of 20, 30, 40, and 50°C to analyse hydration heat evolution. Higher curing temperatures accelerated early hydration but did not proportionally enhance long-term cumulative hydration heat. The presence of coarse aggregates slightly delayed hydration kinetics and increased cumulative hydration heat at later stages, though the observed differences were minimal, making it unclear whether they were due to experimental variability or an actual material effect. A finite element model (FEM) was developed using COMSOL Multiphysics 6.1 to predict the early-age temperature development of in-situ concrete. The heat source for the model was derived from isothermal calorimetry data and adjusted using an Arrhenius-based equivalent age approach to reflect actual hydration heat evolution in concrete. The FEM was validated against semi-adiabatic calorimetry test results. The modelling results underscored the necessity of incorporating real-time ambient temperature variations, as constant-temperature boundary conditions led to discrepancies in predicted temperature profiles. Additionally, models using equivalent mortar hydration heat data overestimated peak temperatures, highlighting the importance of considering coarse aggregate effects. The study also evaluated the applicability of the Three-Parameter Equation (TPE) hydration heat model, originally developed for CEM I, in predicting temperature development in GGBS-containing concrete. While the model provided reasonable accuracy, it consistently overestimated peak hydration temperatures for high-GGBS content mixes, likely due to its assumption of immediate GGBS hydration rather than its delayed activation. Refining hydration models to incorporate the two-stage reaction mechanism of GGBS could improve predictive accuracy. Although this research enhances understanding of early-age concrete temperature development, certain limitations remain. The hydration heat differences between micro-concrete and equivalent mortar were small, making it difficult to determine whether the effect of coarse aggregates on hydration kinetics was genuine or within the range of experimental error. More advanced experimental techniques are required to clarify this issue. Additionally, the FEM was validated under laboratory-controlled conditions, necessitating future field-scale validation to account for real-world thermal interactions. This study enhances the predictive capabilities of hydration temperature models by integrating experimental data with numerical simulations. The findings emphasize the importance of precise boundary condition inputs, improved hydration models for blended cement, and the necessity of incorporating micro-concrete data for accurate temperature predictions. Future research should focus on refining hydration models for GGBS-containing concrete, conducting field-scale validations, and integrating thermal stress analysis to further mitigate early-age cracking risks in concrete structures.