Title: Predicting the Corrosion of Steel in Chloride-Contaminated Concrete

Author(s): Youping Liu and Richard E. Weyers

Publication: Symposium Paper

Volume: 170

Issue:

Appears on pages(s): 379-396

Keywords: Bridge decks; chlorides; corrosion; exposure; modeling; temperature.

DOI: 10.14359/6832

Date: 7/1/1997

Abstract:
A total of 56( 16 indoor and 40 outdoor) simulated bridge deck slabs containing 5 electrically isolated steel reinforcing bars were cast. Study variables were 6 corrosion rates, 2 concrete cover depths, 2 bar sizes and 2 bar spacings. Monthly measured corrosion parameters for the 4 year study period reported here were corrosion current density, concrete temperature at the bar depth and ohmic resistance of the cover concrete. An unguarded (3LP) and guarded (Geocor) device were used to take the monthly measurements. Results demonstrate that the corrosion current density decreases with time at constant chloride contamination levels. Also, the corrosion current density increases with increasing chloride content, decreases with increasing concrete resistance and increases with increasing temperature. Cover depth had little to no influence on the measured corrosion current density. A non-linear model was developed for the 3LP device which demonstrates that the corrosion current density is a function of the water soluble chloride content, temperature at the bar depth, resistance of the cover concrete and the corrosion time after initiation. No similar model could be developed for the Geocor device because of the lack of precision (highly variable instantaneous corrosion current densities) for this device. The predicted corrosion rates are in good agreement with the measured rates. Variability of the 3LP measurements are greater at low corrosion rates than at high corrosion rates.