Resistance of cementitious materials to sulfate attack: quantifying performance with a reliable unidirectional approach

Depending on the environmental conditions, concrete materials can come into contact with sulfate ions which are widely present in rivers, underground water, sewers, seawater and soil. Sulfates can react with the cement paste in concrete and cause damage which impacts the service life of concrete structures. Cement chemistry, transport of sulfate ions, and physical properties of the host material all interact in the complex process of sulfate degradation. Given different conditions of concrete exposure, sulfate attack is commonly divided into "chemical" and "physical" sulfate attack. However, the processes are not fully understood and there is still a need for in-depth investigations of the whole deterioration process. The current standard testing method is adapted for Portland cement systems affected by the "chemical sulfate attack" in submerged conditions. However, only expansion is monitored, which is not suitable for characterizing the increasingly common usage of supplementary cementitious materials in the cement. Moreover, several key factors (e.g., RH & temperature) are usually overlooked so lab tests do not well represent the field conditions. Particularly, the understanding of the damaging mechanism of "physical sulfate attack" is very limited even if different testing approaches have been established. The division into "chemical" and "physical" attack is in itself misleading. In this context, this research project was launched to establish a unidirectional penetration approach for the in-depth comprehensive investigation of joint sulfate attack mechanisms. First, the feasibility of this newly established approach was studied for Portland cement mortars and pastes. A high concentration of sulfate solution and a high water-to-cement ratio was chosen to accelerate the degradation and to obtain results in a reasonable period. Results showed that this new concept of combining chemical and physical sulfate attack mechanisms into a single experiment was viable. Moreover, results showed that using cement paste samples can accelerate the degradation, compared to mortar samples. The sulfate degradation process was more rapid with the capillary rise and water evaporation. A comparison of the macroscopic manifestation of damage (visual appearance and expansion) with the underlying microscopic changes was made considering different key parameters, i.e., water-to-cement ratio, exposure solution concentration, cement type, temperature and humidity conditions (constant or cyclic). Regarding the different cement types, Ordinary Portland cement pastes and sulfate-resisting cement pastes showed higher expansion and cracking, whereas, spalling damage was more pronounced in blended-cement pastes/mortars (slag Portland cement and limestone calcined clay cement). Damage from the chemical attack was more extensive and rapid when high concentrations and cyclic exposure conditions were applied. However, lower water-to-cement ratios leading to lower porosity and/or higher mechanical strength could usually reduce the extent of the damage. Thermodynamic modelling was also conducted with GEMS to better explain the pore solution environment and the link between supersaturation and expansion. The interacting factors of porosity and strength were decoupled. The effect of the curing temperature on the expansion mechanism was also investigated in sulfate-resisting cement mortars under submerged sulfate exposure.