Improved Prediction Models of Creep, Shrinkage, and Relaxation of Modern Concrete

A recent study of the Koror-Babeldoab pre-stressed concrete bridge in Palau revealed a consistent underestimation of the observed structural response due to theoretically incorrect and out-of-date creep and shrinkage prediction equations. In response, excessive bridge deflections were found to be a global phenomenon. Their pervasiveness is independent of geography and design, which enforced the idea that there is a need for a better prediction model to capture the time dependent deformation of concrete. Triggered by these findings, this dissertation work presents a new database of creep and shrinkage measurements, the development of new static creep and shrinkage prediction equations, an improved formulation for the relaxation of concrete, and a discussion and estimation of the potential effects of cyclic creep on bridge structures.;The NU database of over 1500 creep and shrinkage curves, respectively, was developed as a more comprehensive and up-to-date source for the calibration of semi-empirical prediction formulas. Through statistical analyses of these data and theoretical extensions of the existing B3 model, a new creep and shrinkage model, named B4, is developed. The correct multi-decade creep behavior is captured through a joint optimization with the long-term bridge deflection measurements. The B4 model is shown to surpass all existing prediction models in its ability to capture individual test curves as well as a statistical verification and validation analysis. Additionally, the B4 model has the new capability to capture effects of elevated temperatures, the addition of admixtures and additives, and selected aggregates.;As a related development, an improved relaxation function for concrete is formulated which allows for a thermodynamically admissible prediction in the multi-decade regime. Additionally, a suggested fracture mechanics based formulation for cyclic creep implies that the effects of this mechanism are negligible in most bridge designs in comparison to static creep.