Concrete provides a nearly perfect environment for corrosion protection of steel; However, the use of de-icing salts on the highway system has accelerated the deterioration rate of bridge decks in the snow belt. In 1981, over 100,000 bridges were reported to be structurally deficient, and the estimated cost of repair was placed at $93 billion.

Concrete specimens, 1 ft. x 1 ft. x 4 in., containing four pieces of steel reinforcing bars, were prepared. In a first time, five specimens with different rebar networks were cast in order to study the effects of the rebar network on corrosion. Half-cell potential measurements were used to monitor the corrosion behavior of each specimen. In a second time, calcium nitrite, monofluorophosphate and sodium borate were evaluated for their capacity to control corrosion. The various corrosion inhibitors were tested either externally (in the test solution), internally (as a concrete admixture). The specimens were exposed to alternate complete immersion in a 6 wt% sodium chloride solution, plus eventual addition of corrosion inhibitor, for 3 days and 12 hours of drying at 110°F. The effects of corrosion inhibitors were evaluated using half-cell potential measurements and Electrochemical Impedance Spectroscopy (EIS) measurements.

The EIS data were analyzed through a computer assisted EIS data analysis system. This allowed for circuit modeling of the corrosion mechanisms and evaluation of polarization resistance values for the different specimens. In both phases of this work, chloride concentration profiles as function of depth were determined. The half-cell potential measurements in complete immersion were found to give an average over the length of bar or electrically connected bars. The corrosion inhibitors applied internally were found to reduce corrosion better than corrosion inhibitors applied externally.

It was shown that chloride concentration is not the only parameter controlling corrosion initiation.