Date of Award
Master of Applied Science (MASc)
Samir E. Chidiac
Concrete undergoes volume change as it changes phases from plastic to solid. Volume change due to water movement and losses within the concrete are referred to as chemical and autogenous shrinkage and drying and plastic shrinkage are due to water exchange with the surrounding environment. Shrinkage strains need to be investigated as they can have detrimental effects on the serviceability and durability of concrete.
For this study, an experimental program was developed using fractional factorial principles to investigate the effects of curing regime and concrete mixture namely, water to cement ratio (w/c) , water content (w), maximum aggregate size (size), silica fume replacement percent (SF), ground granulated furnace slag replacement percent (GGBFS), and volume of coarse aggregate (CA), on the magnitude of autogenous and drying shrinkage. A new test setup was developed to measure autogenous shrinkage, capillary pressure and temperature. The results were found to concur with those reported in the literature, i.e., moist cured samples exhibit chemical shrinkage and that air cured samples exhibit both chemical and drying shrinkage and that the magnitude of the latter is much greater than the former. Values of drying shrinkage are found to range from 450 to 800 μm/m. The results also revealed that all the parameters studied do contribute to shrinkage but not to the same degree. An increase in the volume and size of coarse aggregate is found to produce concrete that exhibits less drying shrinkage strains. The addition of SF as cement replacement is found in general to increase shrinkage strains. The statistical investigation has revealed that the following parameters, CA volume, w/c2, CA2, w/c*SF, w/c*GGBFS, size*SF, size*CA, w/c*w*size, w/c*SF*GGBFS, and w/c*SF*CA are statically significant to a 90% confidence level.
For autogenous shrinkage, w/c is found to be a significant parameter. The results also revealed that increasing the amount of chemical admixtures, WRA and VEA, has led to a significant increase in strains. Autogenous strains were found to occur when there is a rise in capillary suction pressure, occurring due to self-desiccation.
Seven models proposed in the literature to estimate strains due to shrinkage were evaluated using the experimental data. The majority of these models have been adopted by North American, European or Japanese concrete standards. The assessment has revealed that only two models, namely B3 and ACI - 209 are somewhat adequate in their predictions of strains in concrete that is 28 days or older. Regression models developed in this study are found to provide a better estimation of the concrete shrinkage strains at 3 days, 7 days, 14 days, 28 days and 119 days.
Khairallah, Rabih S., "Analysis of Autogenous and Drying Shrinkage of Concrete" (2009). Open Access Dissertations and Theses. Paper 4546.
McMaster University Library