Effect of Test Methods on Corrosion Phenomena of Steel in Highly Resistive Concrete Systems and Data Interpretations

Abstract

Fly ash and limestone calcined clay cement (LC3) are being used in concrete to enhance chloride resistance. In this study, 60 specimens (with steel in three separate binder systems, namely 100% ordinary Portland cement [OPC], 70% OPC + 30% fly ash, and LC3 with surface resistivity of ≈10 kΩ·cm, ≈25 kΩ·cm, and ≈200 kΩ·cm, respectively) were subjected to impressed corrosion, and the results were compared with 15 lollipop steel-mortar specimens subjected to natural corrosion under a wet-dry chloride environment. It was found that the traditional method of impressed corrosion tests can induce microstructural changes in highly resistive concrete cover and at the steel/concrete interface; hence, these methods are not suitable for evaluating corrosion resistance (such as corrosion rate and corrosion-induced cracking) in highly resistive concrete systems. Further, the Raman spectra from the corroded steel surfaces indicated that the impressed corrosion and natural corrosion tests led to different forms of corrosion (i.e., uniform and pitting, respectively) and different compositions of corrosion products (i.e., α-Fe2O3 and β-FeOOH phases). This led to different expansive stresses, making the lab-to-field correlations inappropriate in the case of highly resistive concrete systems. This paper recommends natural corrosion tests exposed to wet-dry conditions and not the impressed corrosion tests for assessing corrosion phenomena of steel in highly resistive concrete systems.