Effect of shrinkage and load-induced cracking on water permeability of concrete

Abstract

Research on the durability problems of concrete structures has focused on the efficacy of concrete as a barrier against aggressive environments. Research laboratories have mostly used permeability tests on intact concrete for assessment and effectively ignored in-service cracking such as shrinkage and load-induced cracks. This study is an attempt to isolate the effect of shrinkage and load-induced cracks on the apparent water permeability. Shrinkage cracking was achieved by rapid oven drying concrete specimens at 105 C; load-induced cracking was achieved by loading 200 x 100-mm diameter specimens in uniaxial compression levels to stress of 0.3, 0.5, 0.7, and 0.8 of the average cylinder crushing strength. Water permeability was chosen as a measure of the barrier quality. Experimental variables that included five different mixes yielded standard deviations of nominally identical samples that exceeded apparent differences in behavior attributed to the imposed stress. Significant differences in coefficient of permeability were obtained between virgin and oven-dried and resaturated specimens. These results support the view that orientation and distribution of shrinkage cracks were effectively isotropic, whereas load-induced cracks were strongly oriented and localized but did not have a measurable effect on the permeability of specimens that were unloaded when tested, thus indicating significant elastic recovery of the cracked concrete, which caused cracks to close.