Effect of Relative Humidity on Cement Paste: Experimental Assessment and Numerical Modelling

In this study, the objective was to verify whether or not the classical Richard-Fick model was able to take into account the size effect on drying. First, mass loss experiments are made on ordinary cement paste cylinders of 3.6 × 18 cm geometry that enable us to calibrate the model parameters. Second, the quality of identification and drying model was checked by predicting mass loss evolution of small prism of 1 × 5 × 10 mm size dried at different steps of relative humidity. The results demonstrate that the present drying model is able to predict the drying of specimen for different sizes and levels of humidity at ambient temperature. And finally, drying shrinkage experiments for different rates of drying are performed on cement paste cylinders of 3.6 × 18 cm; then, the prediction of drying shrinkage evolution was made using a simple model where the drying shrinkage is supposed to be linear with respect to the relative humidity. The result appears to be satisfactory. Since the spatio-temporal evolution of water content is needed as input to shrinkage prediction, the latter result confirms that the identification of drying model parameters is trustworthy.