Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article


Doctor of Philosophy


Civil and Environmental Engineering


Moncef L. Nehdi


Self-healing efficiency of cement-based materials has so far been evaluated mostly through the healing of surface cracks, without adequately capturing the dominant effects of environmental exposure or accurately quantifying the volume of cracks healed. In addition, the effects of diverse additions such as silica-based materials, swelling agents, superabsorbent polymers, and carbonating minerals on self-healing performance under different environmental exposure, remain largely unexplored.

In this dissertation, multiple test methods were used to investigate self-healing of cracks in cement mortar incorporating metakaolin, bentonite, fly ash, superabsorbent polymers, and calcium carbonate microfiller under different environmental exposure (i.e. cold and hot temperatures, high and low humidity, wet and dry cycles, and continuous underwater submersion). Change in crack width was monitored using optical microscopy. Scanning electron microscopy coupled with energy disperse X-ray analysis was used to identify healing compounds. Mercury intrusion porosimetry and water absorption were employed to assess porosity. X-ray computed micro-tomography (X-ray µCT) with 3-dimensional image processing was used to segment and quantify cracks before and after healing. The findings should stimulate concerted research efforts to bridge the gap between ideal laboratory conditions and realistic field exposure in future self-healing research endeavors.

Furthermore, an attempt was made to develop a hybrid artificial intelligence-based model to accurately predict the ability of concrete to heal its own cracks. A comprehensive database of concrete crack healing was created and used to train the proposed GA–ANN model. The results showed that the proposed GA–ANN model can capture the complex effects of various self-healing agents (e.g. biochemical material, silica-based additive, expansive and crystalline components) on the self-healing performance in cement-based materials. This could allow tailoring self-healing strategies for enhancing the durability design of concrete, thus leading to reduced maintenance and repair costs of concrete civil infrastructure.

Summary for Lay Audience

Concrete is the second most used material on earth after water. The vast majority of the built civil infrastructure (i.e. bridges, tunnels, dams, etc.) is made of concrete. Although it is cost-effective and able to carry relatively high compressive loads, concrete is vulnerable to cracking. Therefore, harmful substances can easily penetrate into the concrete matrix, leading to premature damage. In addition, periodic inspection and maintenance of concrete structures is time consuming and often not effective.

Recently, inspired by the healing process of wounded skin, the concept of designing self-healing concrete has become an area of great interest. Self-healing of concrete implies that, without human intervention, cracks in concrete structures can automatically be filled. If successful, this could save billions of dollars in maintenance and repair costs, helping to build sustainable and more resilient infrastructure.

Despite the advent of abundant literature on the self-healing of concrete, there is currently lack of information on the effect of environmental exposure on the self-healing of concrete. Therefore, in the present study, the self-healing behavior of concrete incorporating various additives under different environments (i.e. cold and hot temperatures, high and low humidity, wet and dry cycles, and continuous underwater submersion) was investigated. In addition, several techniques such as CT scan, non-destructive mechanical testing, and 3D image analysis were used to evaluate and quantify the extent of self-healing in concrete. Moreover, an artificial intelligence-based model was developed to accurately predict the concrete’s capability to heal its own cracks. This approach could exploit existing experimental findings for developing a predictive tool, which ultimately could form a basis for developing guidelines for the design and prediction of self-healing concrete.