Characterizing Porous Volume of Cement-Based Concrete by Multiscale Image Analysis

Abstract

The porous volume of cement-based concrete is the key to understand its mechanical properties and durability. In this research, we apply X-ray tomography (XRT) and focused ion beam/scanning electron microscopy (FIB/SEM) to scan the multiscale porous volume of concrete with a resolution from 25 μm down to 10 nm. Based on the three-dimensional (3D) images, the pore morphology, pore network representativity, porosity, pore network connectivity, and pore size distribution (PSD) in different scales are analyzed and further discussed. Results show one sample can hardly be representative for observed zones, especially for the FIB/SEM and XRT1 samples. Thus, three samples are applied to FIB/SEM and XRT1 imaging, which reduce the relative error of porosity from 31%–33% in the micron scale and 15%–36% in the nanoscale to 18%–19% and 9%–21%, respectively. The combination of porosity in a multiscale has a similar value with that measured by mercury intrusion porosimetry (MIP). Meanwhile, the combined PSD has a similar curve form and peak size to that of MIP, which implies that the combination of image analysis in a multiscale provides a preliminary view of a multiscale porous volume of concrete.