Autogenous Shrinkage of Cementitious materials containing BFS

Abstract

In the Netherlands blast furnace slag (BFS) cement is commonly used. Concretes with this type of cement are known for good durability properties. Where concrete is designed to crack in order to let the incorporated steel reinforcement take over tensile stresses, it is mainly for durability reasons that non-designed cracking of concrete is undesired. Percolated cracks may lead to leakage problems or ingress of chemicals, which can cause deterioration of the concrete matrix or reinforcement corrosion. When material deformations are restrained, stresses are induced, possibly leading to cracking. Deformations in the early age are of most interest, with general properties of the concrete material still being developed. For the early age this is mainly caused by temperature and autogenous deformations, due to cement hydration. Importance of autogenous shrinkage has been recognized for high performance concretes, having relatively low water to cement ratios. It is becoming known that autogenous shrinkage is higher for materials containing BFS. As of late striking results have been found for magnitudes of autogenous shrinkage for common normal strength concretes with water to binder ratios > 0.4. This thesis gives an overview on the topic of autogenous deformation for normal strength concretes from practice, containing BFS. First, results from previous research have been verified, followed by an indication of practical relevance of the topic. Main focus is on the influence of different material constituents, by changing the proportions or type of material. Experiments are conducted on concrete and cement paste. Magnitude of autogenous shrinkage is related to total shrinkage and compared to prediction models known in practice. Combined with results on general properties an indication is given for the possibility of cracking due to autogenous shrinkage. Experiments on concrete regard influence of the water to cement ratio, aggregate type and addition of polypropylene fibres. Measurements start after one day and run for 197 days. Up to five specimens are used per test, which are either kept in autogenous conditions or are exposed to drying from 3, 7 or 28 days. For experiments on cement paste a relatively new device is used, called the Mini-TSTM. Recommendations are given for on-going development, but tests show promising results. It is possible to measure either free deformation or induced stress in case of restrained deformation. Cement pastes are used to determine influence of cement and filler type and content. Furthermore information is given on shrinkage development starting from the moment of initial set. Results on free deformation and self-induced stress under autogenous conditions are compared and show a reduced tensile strength when deformations are restrained. Results for tests on concrete show an ongoing shrinkage. From tests on cement pastes an expansion is found on the first 2~3 days. Combination of both results estimates the magnitude of tensile stress inducing shrinkage. Development of shrinkage for cement paste with BFS shows to be substantially different from that of ordinary portland cement paste, it is mainly steeper and of higher magnitude. The type of cement seems to be the main factor influencing the magnitude of autogenous shrinkage. Measured total shrinkage relates well to the values from prediction models used in practice, but for the use of BFS cement (CEM III/B) subdivision into autogenous and drying shrinkage does quite differ. Autogenous shrinkage is found to be half of the total shrinkage. At 28 days the magnitude of autogenous shrinkage approximates the maximum strain at break. The main recommendation is to not neglect autogenous shrinkage when BFS cement is used, even for normal strength concretes commonly used in practice.