Modeling chloride transport in cracked concrete: a 3-D image–based microstructure simulation

The prediction of the service life of concrete materials is difficult, mainly because of their complex heterogeneous microstructure and their random nature. Studying the presence of cracks in concrete and their effect on chloride transport and binding properties is of great interest in civil engineering. Cracks with different widths and depths will reduce the effective cover thickness and accelerate the transport of chloride ions. So, it is highly desirable to develop a model predicting the chloride diffusion depth in cracked concrete while considering the real microstructure including cement paste, voids, and aggregates. While current models consider concrete at various levels of complexity in predicting the initiation of chloride-induced corrosion, considering the influence of cracking is generally beyond their scope. In this study, a 3-D image-based microstructure simulation procedure was developed to model the chloride ingress in cracked concrete. A micro-X-ray fluorescence (XRF) test was conducted to measure the chloride concentration profile of a concrete sample. The notched concrete sample was put in a chloride ponding test for 30 days before the micro-XRF measurement. A 2-D simulation result, with a mesh based directly on the XRF characterization of microstructure, showed good agreement with the micro-XRF measurement. With this validation, two different 3-D concrete microstructures were generated and meshed in 3-D and a commercial software package was used to accurately compute the influence of cracking on chloride diffusion with binding. The chloride concentration gradient in the crack changed the concentration profile along the crack and nearby irregular aggregate surfaces continuously. Comparison to micro-XRF measurement data indicates that the contributions of the crack play a significant role in the chloride ingress.