Effect of blast furnace slag on chloride penetration into concrete bridges

In the present paper, the results of an experimental study are presented, where the objective was to provide more information about the effect of granulated blast furnace slag on the resistance of concrete against chloride penetration. In a concrete with a water/binder ratio of 0.40, the portland cement were successively replaced by 40, 60 and 80% of slag, the effect of which was tested by observing the change of both chloride diffusivity and electrical resistivity. For the testing of chloride diffusivity, an accelerated non-steady state migration test method was applied, while the electrical resistivity was tested by use of both the two electrode and the four electrode method. At an age of 3 days, all the slag containing concretes showed a significantly higher diffusivity than that of the pure portland cement, but already after 7 days, the slag concretes showed a distinctly lower diffusivity compared to that of the pure portland cement concrete. For all the slag concretes, the reduction in chloride diffusivity was very rapid compared to that of the portland cement concrete, and the more slag the lower the diffusivity. By 28 days, the chloride diffusivity was reduced from 11.2 to 4.9, 3.6 and 2.3 × 10 ^-12m²/s, respectively, while after 365 days, a chloride diffusivity of 3.0 to 1.0 × 10^-12m² for the slag concretes compared to 7.0 × 10^-12m²/s for that of the pure portland cement concrete was observed. Based on the two electrode method, the electrical resistivity of all the slag containing concretes increased very rapidly compared to that of the pure portland cement, and the higher the slag content the higher the resistivity. Thus at 28 days, the resistivity of the slag concretes varied from 10 to 30 kohm·cm compared to approximately 5 kohm·cm for that of the pure portland cement. After 365 days, the resistivity varied from 23 to 50 kohm·cm for the slag concretes compared to approximately 8 kohm·cm for the portland cement concrete. Based on the four electrode method, the slag concretes increased even more rapidly than that obtained by the two electrode method. Thus at 28 days, the resistivity of the slag concretes varied from 22 to 58 kohm·cm compared to approximately 8 kohm·cm for that of the portland cement, and after 365 days, the resistivity varied from 38 to 83 kohm·cm for the slag concretes compared to approximately 12 kohm·cm for the portland cement concrete. Although the observed levels of resistivity obtained by the two test methods were quite different, the results showed a good correlation. The obtained test results also showed a good correlation between chloride diffusivity and electrical resistivity, which is in accordance with both existing experience and theory. Thus, after establishing the relationship between the chloride diffusivity and the electrical resistivity for a given concrete with given temperature and moisture conditions, a regular monitoring of the electrical resistivity can be used as a basis for performance-based concrete quality control of chloride diffusivity during concrete construction. In order to better demonstrate how the observed effect of slag would affect the risk of steel corrosion for a given type of concrete structure in a given type of marine environment, some probability-based durability analyses were carried out. Based on a 10% probability of corrosion as a serviceability limit state, the pure portland cement type of concrete would exceed such a limit state already within a period of 10 years, while for increasing incorporation of slag, a substantially longer period would be reached, and the more slag the longer the period. Thus, for a slag content of 60%, a 10% probability of corrosion would not be reached within a period of 100 years.