Evaluation of Corrosion Protection Systems and Corrosion Testing Methods for Reinforcing Steel in Concrete

Abstract

Corrosion protection systems for reinforcing steel in concrete and the laboratory methods used to compare these systems are evaluated. The systems evaluated include concrete with a low water-cement ratio, two corrosion inhibitors (Rheocrete 222+ and DCI-S), three microalloyed Thermex-treated steels, one conventional Thermex-treated steel, MMFX microcomposite steel, epoxy-coated steel, two duplex steels (2101 and 2205), and three heats of uncoated normalized steel, used as control specimens. The duplex steels were tested in both “as-rolled” and pickled conditions. The corrosion protection systems are evaluated using the rapid macrocell, Southern Exposure, and cracked beam tests. Some corrosion protection systems are also evaluated using the ASTM G 109 test. The corrosion rate, corrosion potential, and mat-to-mat resistance are used to compare the systems. An economic analysis is performed to determine the most cost-effective corrosion protection systems. The degree of correlation between the Southern Exposure, cracked beam, and rapid macrocell tests is determined. The coefficient of variation is used to compare the variability in the corrosion rates and the total corrosion losses obtained using the different test methods. Impedance spectroscopy analysis is performed to obtain equivalent electrical circuits to represent the rapid macrocell and Southern Exposure tests. Results show that microalloyed steel and conventional Thermex-treated steel show no improvement in corrosion resistance when compared to conventional normalized steel. In mortar or concrete with a low water-cement ratio, corrosion losses are lower than observed at higher water-cement ratios for either cracked or uncracked mortar or concrete. In uncracked mortar or concrete (rapid macrocell and Southern Exposure test) containing corrosion inhibitors, corrosion losses are lower than observed at the same water-cement ratio but with no inhibitor. For concrete containing inhibitors, with cracks above and parallel to the reinforcing steel (cracked beam test), Rheocrete 222+ improves the corrosion protection of the steel, while DCI-S does not. iii MMFX microcomposite steel exhibits corrosion losses between 26 and 60% of the losses of conventional steel. Based on corrosion potentials, the two steels have a similar tendency to corrode. MMFX steel has a higher chloride corrosion threshold than conventional steel. Epoxy-coated steel, intentionally damaged by drilling four 3.2-mm (1/8-in.) diameter holes in the coating, exhibits low corrosion losses based on the total area of the bar, between 6 and 19% of that of uncoated conventional steel. Pickled 2101 and 2205 duplex steels exhibit very good corrosion performance. The average corrosion losses for these steels ranged from 0.3 to 1.8% of the corrosion loss for conventional steel, and in most cases, the corrosion potentials indicated a very low tendency to corrode, even when exposed to high chloride concentrations. 2205 steel performs better than 2101 steel when tested in the same condition (pickled or nonpickled). For bars of the same type of steel, pickled bars exhibit lower corrosion rates than the bars that are not pickled. Based on present cost, decks containing pickled 2101 or 2205 steel are more cost effective than decks containing epoxy-coated or uncoated conventional steel. Results from the rapid macrocell, Southern Exposure, and cracked beam tests show good correlation in most cases, and have similar variability in corrosion rates and losses. In general, total corrosion losses have less variability than corrosion rates.