Abstract
When concrete is heated under a sustained load, a large amount of shrinkage occurs to compensate for thermal for expansion. This shrinkage was termed the load-induced thermal strain (LITS), and must be considered when performing a deformation analysis of reinforced concrete structures that are subjected to fire. In the constitutive strain models used in such analyses, the LITS is a function of the concrete temperature. However, the relationship between the LITS and temperature is influenced by the water of the concrete, making it difficult to apply these models to concrete with different water conditions. In the present study, the influence of the curing conditions (air-dried, oven-dried and sealed curing) and water/cement ratios (40%, 50% and 65%) on the LITS was investigated based on results obtained from the transient tests for measuring total thermal strain of concrete specimens. The relationship between the LITS and specimen weight loss was also evaluated in order to develop a comprehensive model for the LITS that is appreciable to concrete with different curing conditions and water/cement ratios.
In the transient tests for measuring total thermal strain, specimens ware heated at a constant rate of 1.5°C/min. up to 800°C under constant load. The ratio of the constant load to compressive strength of air-dried specimens at room temperature was settled between 0.1 and 0.7. The LITS was calculated by subtracting value of the free thermal strain from the total thermal strain. The heating rate in the transient tests for measuring the weight loss was same as the tests of the total thermal strain.
The main results obtained from the present study are as follows:
(1) The influence of the water evaporation on the LITS was evident between 100°C and 200°C.
(2) The influence of the dehydration and decomposition of the cement hydrate on the LITS was evident above 200°C.
(3) The LITS due to dehydration and decomposition was larger than that due to water evaporation.
(4) The relationship between the LITS and load level was seen to be linear, regardless of the curing conditions or water/cement ratio.
(5) The numerical model for the LITS was proposed as a function of the weight loss of the specimens. And, it was indicated that the proposed model could be appreciable to concrete with different curing conditions and water/cement ratios.
