Next: Experimental Procedures Up: Main Previous: Main

Introduction

As with any material, an understanding of the links between the microstructure of cement-based materials and their properties is needed to allow the design of systems with improved performance. Unfortunately, three-dimensional quantitative characterization of microstructure is extremely tedious and difficult. Although two-dimensional scanning electron microscopy (SEM) images of hydrated cement paste are straightforward to obtain, it is the three-dimensional microstructure that often has a critical influence on properties. Unlike phase volume fractions that are statistically the same in two and three dimensions for isotropic systems, the connectivity or percolation of phases is vastly different in two and three dimensions. Because percolation aspects are critical to the mechanical and transport properties of cementitious materials, 1, 2 a three-dimensional representation of microstructure is required. Although difficult to obtain experimentally at the necessary resolution, such representations can be simulated using computer models. Using computer models to represent the microstructure of cement-based materials has evolved significantly over the past 10 years or so, as reviewed in Panel A. In the present study, an experimental validation was conducted to determine the relationship between model cycles and real time for two cements at three different water-to-cement (w/c) ratios based on measurements of degree of hydration via nonevaporable water content, heat release, and chemical shrinkage. In addition, temperature effects and curing under sealed conditions were assessed experimentally for incorporation into the computer model. The overall model and experimental program are summarized in the flow diagram in Fig. 1.

Fig. 1. Flow diagram summarizing experimental and modeling program for predicting cement performance.


Next: Experimental Procedures Up: Main Previous: Main