Over the past several decades, many researchers have emphasized the prediction of the early-stage strength gain of concrete. Such a prediction is of special importance for cold-weather and hot-weather concreting and for massive concrete members. In spite of extensive research done over the past decades, there are still several problems in quantitatively predicting the strength development at both early and later ages.
In this experimental and analytic research for strength prediction of early-age strength, the strength development for various curing histories was investigated with a particular regard to the influences of curing time points with given temperatures. For this purpose, four different points of curing time were considered with an individual interval of 24 hours. Two different temperatures of 5℃ and 40℃ were applied for the selective intervals while the rest period days were under the reference curing condition of 20℃.
A new model for the strength prediction was suggested based on the rate constant model. In this model, the equivalent ages introduced in the Saul and Arrhenius models were modified in order to show the effects of curing temperature at different ages. Test results show that the concrete subjected to a high temperature at an early-age attains higher early-age strength but eventually attains lower later-age strength. The concrete subjected to a low temperature at early age leads to lower early-age strength but almost the same later-age strength. Moreover, the proposed model showed better agreement with the test results than the existing models.
A strength prediction model to improve the later-age strength with curing temperatures was proposed. In order to verify the proposed model, the existing data which had been reported was used. The effects of diffusion barrier with curing temperature during hydration on later-age strength were considered using rate constant model. The verification of the model was performed by regression analysis, in which the existing data were converted to the relative ratio of 28-day strength with 8 average curing temperatures in a range of -0.6~59.7℃. It was found that the temperature influence factors, such as rate constant, limit strength and reaction exponent were expressed in function of curing temperature. The results of regression analysis showed that the proposed model was higher in reliability than the previous model. Especially, the proposed model had the higher accuracy at later-age meanwhile it made no significant difference from previous model at young-age.
