Freeze-thaw de-icer salt scaling resistance of concrete containing GGBFS

The use of ground granulated blast furnace slag (GGBFS) as partial cement replacement improves the durability performance of concrete against, sulphate attack, alkali silica reaction and reinforcement corrosion compared to ordinary Portland cement (OPC). Improved durability performance is however not observed for the laboratory salt scaling resistance of concrete containing high replacement percentages. Although there are several existing theories used to describe the scaling degradation mechanism, none of them explain the influence of GGBFS on scaling performance.;The primary research goal is to determine the cause(s) for the influence of GGBFS on the scaling performance of concrete when tested in accordance with MTO LS-412. This research program focused on investigating the effects of replacing OPC with GGBFS on the mechanical and mass transport properties, hydration kinetics, chemical processes, and freeze-thaw durability performance of dry cast concrete. The primary variables considered in the experimental investigations include the water-to-binder ratio (0.31 and 0.38), percentage of GGBFS (0 to 60), total binder content (270-450 kg/m3), and age (28 days and 2 years).;Two multi-variable statistical regression analysis models were developed to identify the variables which affect scaling performance. The analysis considered various concrete mixtures and sample preparation variables. The statistical models captured approximately 70% of the variability in the experimental scaling performance data. Accordingly, investigating the effect of other factors such as hydration kinetics and chemical interactions with cement phases on scaling resistance of concrete containing GGBFS was warranted.;The role of hydration kinetics on the development of the compressive strength, hardened density, ultrasonic pulse velocity (UPV) measurements and its implications on scaling damage were evaluated. At 28 days, concrete containing up to 60% GGBFS exhibited equal or superior mechanical properties as compared to OPC. This finding implied that the strength was not a contributing factor, as suggested by others, to the poor performance of concrete containing high GGBFS when subjected to the MTO LS-412 laboratory scaling resistance test at 28 days.;The microstructure of concrete containing GGBFS is affected upon exposure to chloride solution owing to chloride binding processes. The percentage physical and chemical binding are similarly equal at a chloride concentration of 0.5 M for mixtures containing 0 and 40% GGBFS. At chloride concentrations greater than 1 M, the total bound chloride is predominantly chemically bound for the 40% GGBFS mixture. After exposure to 0.5 M solution, a reduction in total porosity and the refinement of the capillary pores was observed and the changes to the pore characteristics is attributed to the combined effect of the concrete's pore size distribution and the formation of precipitate from chemical chloride binding reactions which modify the capillary pores.;Results revealed that the cementing material proportions affect changes to the concrete's near surface microstructure which affect the capillary action at the concrete's surface and are correlated to scaling damage to concrete containing high percentages of GGBFS. The liquid uptake is determined using an adapted capillary suction force model that includes the interplay between porosity, chloride binding capacity, and degree of hydration of the concrete. The capillary suction model was found to correlate with experimentally measured data, namely, a new property, 'ionic sorptivity', which provides a measure of the difference between the capillary uptake between water and saline solution. Ionic sorptivity reflects the degree of saturation of the concrete which controls the development of hydraulic pressure that induces freeze-thaw damage. An indicator for scaling resistance, freeze-thaw salt scaling resistance factor (F/T SSRF), which includes the degree of hydration, ionic sorptivity and tensile strength of the concrete, is proposed for evaluating the scaling performance of concrete containing various factors including percentage of GGBFS, binder content, w/b ratio and age. The F/T SSRF is found to correlate well with the mass loss thus providing the same pass/fail prediction stipulated in MTO LS-412.