Measurement of entrained air-void parameters in Portland cement concrete using micro X-ray computed tomography

Project Details









Lu, Haizhu; Peterson, Karl; Chernoloz, Oleksiy


University of Arizona College of Engineering


Air content, Air entrained concrete, Air voids, Concrete pavements, Freeze thaw durability, Highways, Image processing, Materials, Pavements, X-ray computed tomography

Project description

The entrained air-void system in concrete is closely related to freeze-thaw durability in concrete pavements or other structures. For either research or forensic purposes, reliable and economical methods for the quantification of entrained air are desirable. This study explores the potential of using micro X-ray computed tomography (μCT) to measure entrained air-void parameters in concrete. A series of small cores (6 mm dia.) were retrieved from larger (100-mm-dia.) cores from two different concrete pavements, representing both adequate and marginal air contents, and scanned at a resolution of 7.5 μm/pixel. A systematic procedure based on image processing is proposed to address practical difficulties such as void/solid thresholding, air-type discernment (entrained air-voids vs. voids in aggregate) and the separation of bubbles within close proximity to each other (e.g. clustered air-voids). Air content and specific surface were measured directly from the three-dimensional (3D) reconstructed X-ray images, while values for paste content were derived from manual point counts performed on two-dimensional (2D) slices obtained from the 3D images. The derived values for air content, specific surface and paste content were used to calculate Powers’ spacing factor. To assess the issue of local fluctuations of material constituents and the limited dimensions of the small cores, uncertainty associated with the sample volume of concrete under measurement was also estimated. Based on the results in this study with regard to the work involved in sample preparation, data analysis and uncertainty bounds, μCT has been found to be a viable option for measurement of spacing factor and specific surface, but due to limitations imposed by the dimensions of the sample size (6-mm-dia. cores), the method is not appropriate for bulk air content determination.