To achieve optimal performance from cement-based materials in the field, proper curing conditions must be maintained throughout the first few weeks of their life [1,2]. Unfortunately, due to a lack of quality control in the field, concretes and mortars often experience considerable drying before the cement paste matrix has undergone "complete" hydration. For some materials, such as the thin cement/polymer composite mortars used as rendering materials, later addition of liquid water to the dried composite is not able to "reinitiate" the cement hydration process . For others, this re-initiation of hydration causes significant internal damage to the microstructure and a loss of mechanical properties . A basic understanding of water movement and the kinetics of both hydration and drying in cement-based materials is thus needed to produce systems with adequate performance. This applies to freshly cast cementitious systems, curing compounds, and repair materials. In this paper, experimental and computer modeling studies are applied to elucidating an understanding of the basic mechanisms of water movement in fresh cement pastes and mortars.
For the experimental studies, use was made of an X-ray environmental chamber that has recently been constructed at the Technical University of Denmark to examine building materials exposed to various drying environments . Both relative humidity and air temperature can be controlled within the chamber, which also serves as a shield from the X-ray source. In this study, the X-ray environmental chamber was used to monitor water movement during the drying of small cement paste and mortar specimens. Because the X-ray absorption is proportional to the density of the materials through which the X-rays are passing, a high water-to-cement ratio (w/c) less dense paste will absorb less X-rays than a lower w/c paste. Similarly, a paste which has dried out will absorb less X-rays than the same paste under its initial saturated conditions, allowing a rapid non-destructive quantification of the moisture distribution within the drying/hydrating sample.
In addition to the experimental studies, the NIST three-dimensional cement hydration and microstructure development computer model [6,7], CEMHYD3D, has been extended to directly consider drying and the consequences of empty (as opposed to water-filled) porosity on microstructure development and hydration kinetics. Based on the experimental observations, appropriate algorithms were developed for simulating the drying of fresh cement paste and the extended code was applied to simulating several of the systems on which experimental measurements were made [8,9].