The early age short-term creep of hardening cement paste: load-induced hydration effects

Abstract

Short-term creep and shrinkage strains were monitored at early age on hydrated Portland cement pastes prepared with two different water-cement ratios (0.35 and 0.50). Creep experiments were conducted in an environmentally controlled chamber maintained at (96 ± 2)% relative humidity. The three ages at loading investigated were 18, 24 and 30 hours. The stress-strength ratio applied to the specimens at the age of loading, using a miniature loading system, was 0.30. Cement paste specimens were in the form of "Tshaped" columns with a minimum thickness value (for the web and flanges) of less than 1.2 mm. Load-induced hydration (not normally considered in creep prediction) of normal strength cement paste (w/c=0.50) was found to occur at early times following the loading at 18 hours. An analytical model was developed in order to predict the creep coefficient of normal and high strength cement pastes from early age data. The model accounts for load-induced hydration effects.

Microstructural Modeling of Early-Age Creep in Hydrating Cement Paste

“…7. Results of creep compliance in the cement pastes for three different ages of loading; L18, L24, and L30 denote loading from 18, 24, and 30 h, respectively, with constant density of C-S-H, whereas L18 D, L24 D, and L30 D denote loading from 18, 24, and 30 h, respectively, considering the effect of C-S-H densification; the experimental data reported by Tamtsia et al (2004) were performed using the compressive stress-strength ratio of 0.3 applied to the specimens at the age of loading processes of microstructural development and creep, which occur in parallel, could be captured and their combined effect simulated in these models.…”

“…The input of cement mixture for the simulations is based on the experimental data from Tamtsia et al (2004) with which the computed creep results will be compared. Based on the composition of the main minerals with the assumption of 5% gypsum content, the Bogue cement composition is C 3 S (55%), C 2 S (20%), C 3 A (10.5%), and C 4 AF (9.5%).…”

“…The hydration degrees in the simulations are calibrated to match the profiles of hydration degree (at a constant relative humidity of 96 AE 2% and a constant temperature of 22 AE 2°C) as obtained by Tamtsia et al (2004) from thermogravimetric analysis characterizing the state of water in the samples. Tamtsia et al (2004) claimed that a sustained load for loading from 18 h applied on cement samples during creep tests may accelerate the hydration rate, which in turn influences the microstructural development and the creep properties.…”

“…Tamtsia et al (2004) claimed that a sustained load for loading from 18 h applied on cement samples during creep tests may accelerate the hydration rate, which in turn influences the microstructural development and the creep properties. However, it is shown from the results in Tamtsia et al (2004) that the hydration degree profiles are very similar for the loaded and unloaded samples for both ages of loading at 24 and 30 h, and hence creep loads seem to have no effect at all on the hydration kinetics. Therefore, the hydration degree profile from the measurement for the case of loading at 24 h is used to obtain the evolution of the digital hydrating microstructures for the creep simulations of loading at both 18 and 24 h.…”

“…For validation with experiment, the inputs of the simulations are based on the data set for a cement paste at water to cement ratio of 0.5 from Tamtsia et al (2004) with which the outputs of the creep simulations at three different ages of loading are compared. Due to scarcity of experimental data for early-age creep of cement paste, only this data set has been found to be sufficient for validation of the simulations.…”