Modeling of autogenous deformation in cementitious materials, restraining effect from aggregate, and moisture warping in slabs on grade

Autogenous shrinkage in concrete is a consequence of self-desiccation from cement hydration. Internal pore-stresses develop as the hydration progresses. Pore-drying and thus autogenous shrinkage increases with extent of hydration (or time) and is especially pronounced in systems with low water-cementitious ratios. The risk of early-age cracking due to restrained autogenous shrinkage is especially a concern in bridge-deck concrete. There is very little data available on autogenous deformation due to the challenges in measurement which requires sealed curing conditions and low frictions between the specimen and the surrounding. The thesis focus has been on the measurement and subsequent modeling of the autogenous deformation development in cementitious systems containing various amounts of supplementary cementitious materials. A new methodology is presented for accurate measurement of this property. New prediction models are developed for both cement paste and concrete shrinkage by incorporating a shrinkage-stress equilibrium model developed by Pickett with a time-domain model developed by Freiesleben-Hansen and Pedersen.;Finally, the cause of extensive moisture warping found in Jointed Plain Concrete Pavements (JPCPs) is explained through combining drying shrinkage and autogenous moisture predictions.;The major findings are: (1) Autogenous shrinkage of cement paste is a result of the porous hydration products (PHP), the majority of which is calcium silicate hydrate (C-S-H) gel, combined with a reduction in pore humidity within the hydration products. (2) Cementitious blends of portland cement and ground granulated blast-furnace slag (GGBFS) develop initially lower autogenous shrinkage, proportional to the reduction in portland cement, while later-age pozzolanic reactions cause increase in porous hydration products and associated pore-drying. Ultimate autogenous shrinkage may or may not increase. (3) Concrete autogenous shrinkage is effectively modeled using the improved Pickett's model as a function of paste shrinkage and aggregate concentration. A new warping theory is introduced for understanding the extensive joint uplift found in the field JPCP slab. This will aid in developing performance-based materials and mitigating shrinkage cracking.