Role of binder, permeability-reducing admixtures and cracking in the watertightness of concrete structures

Publication Type:
Thesis
Issue Date:
2017
Full metadata record
Files in This Item:
Filename Description Size
01front.pdf339.47 kB
Adobe PDF
02whole.pdf12.06 MB
Adobe PDF
In the absence of other deterioration mechanisms, concrete structures can be deemed to be durable and serviceable when they resist water penetration and perform their function with minimal maintenance during their service life. Watertightness of in-service concrete structures is assessed by evaluating the water penetration properties of the concrete. The main sources of water penetration in a concrete structure may arise from the concrete matrix and the presence of cracks, joints and construction defects in the concrete elements. In this study, the water penetration through the uncracked and cracked concrete matrices is investigated. The water penetration through the uncracked concrete matrix is influenced by different parameters such as mix constitutions including the addition of the chemical admixtures, and the placing and curing conditions of the concrete. The popularity of permeability-reducing admixtures (PRAs) in producing watertight concrete has increased significantly in the last couple of decades due to their cost and time efficiency compared to the application of membrane-based waterproofers. However, the available information with regard to their performance has not been expanded at the same pace, so the reliance is mostly based on the manufacturer’s data associated with experimental evidence. This study aims to evaluate the effect of common PRAs as well as the binder type, binder content, w/b ratio and maturity of the concrete on the absorption and permeability properties of the concrete. The study undertaken includes experimental research and statistical analyses to reveal the influence of the aforementioned factors and their significance. Results indicate that although the PRAs were beneficial in certain cases, the effect of binder characteristics such as binder content, binder type and w/b ratio was more pronounced than the effect of the PRAs. The efficiency of the PRAs was found to be a function of the w/b ratio, binder type and the type of test method used to evaluate water penetration. From a design perspective, water penetration in concrete structural elements mainly results from the presence of cracks that increase water penetration by several orders of magnitude when compared with the uncracked concrete matrix. In the current study, the effect of cracking on the water penetration properties of concrete was investigated experimentally and numerically. Controlled cracks with predefined widths of 0.1 mm to 0.4 mm were induced in concrete specimens through feedback-controlled splitting tensile tests. The permeability characteristics of cracked concrete specimens were determined and compared with those of uncracked concrete specimens. Crack healing was investigated over time in concretes with and without PRA additions. Numerical relationships were developed between crack width and water penetration. Results indicate that the water penetration in the cracked concrete is largely influenced by the crack size and geometry, and the flow rate is cubically related to the crack width. The induced cracks in all types of concrete evaluated were found to heal under the conditions examined when these crack widths were below 0.1 mm. The cracks in concretes containing crystalline PRAs were found to heal at widths of up to 0.2 mm. To predict the water penetration in concrete structures, finite element analysis was employed to model the permeability of cracked concrete under service loads. The permeability of the concrete specimens was modelled and validated against the experimental results. Subsequently, two common types of watertight concrete structures including a roof slab and a rectangular water tank were analysed. The areas in these concrete structures with high potential for water penetration due to cracking (before and after self-healing) were predicted via finite element models. Results were in agreement with the experimental test data carried on the cracked concrete specimens.
Please use this identifier to cite or link to this item: