ABSTRACT
The investigation was performed on cement pastes of w/c 0.3 and corresponding mortars with 40% aggregate volume fraction.
The internal RH in the cement paste exposed to temperature cycles of 17-23°C was measured using water activity sensors. As a result, the continuous evolution of the coefficient ΔRH/ΔT in the material undergoing self-desiccation was obtained (Fig. 1). As can be seen, the drop of internal RH due to
1 INTRODUCTION
Thermal stresses and the risk of cracking of concrete will depend on the characteristics of the structure (size, degree of restraint) and properties of the material, among which the tendency of the material to deform due to the temperature change will play an important role. The latter property can be quantified with the coefficient of thermal expansion (CTE), e.g. Sellevold & Bjøntegaard (2006). It has long been observed that the CTE increases during maturing of the material and after further drying to the environment. Sellevold & Bjøntegaard (2006) suggested that the total thermal expansion in partially saturated concrete origins from the following mechanisms: pure thermal dilation of constituents, thermal shrinkage (manifested in delayed contraction as temperature is increased, which is due to the redistribution of water from gel to capillary pores) and RH changes due to temperature changes. This work deals in particular with the latter mechanism. The first observation is that in capillary porous media as concrete the internal RH change due to the unit change of temperature, expressed with a coefficient ΔRH/ΔT, takes positive values, i.e. the increase of temperature leads to an increase of RH, and in consequence the resulting deformation occurs in the same direction as the immediate part of the pure thermal dilation. This effect of changing RH becomes more apparent as RH and saturation
the self-desiccation process is accompanied by the increase of ΔRH/ΔT. The values of the ΔRH/ΔT coefficient presented in Figure 1 (more than 0.70%RH/°C at 7 days) are considerably higher than those reported by e.g. Selevold & Bjøntegaard (2006) which did not exceed 0.15%RH/°C at 90% RH. As shown in the following section, the high values can be validated with the CTE tests.
