A time to cracking model for critically contaminated reinforced concrete structures

TR Number
Journal Title
Journal ISSN
Volume Title
Virginia Tech

In addressing the need for a site specific corrosion model for concrete structures, Bazant’s time to first cracking model was used as the basis for an experimental research study. The theoretical model, which expresses the time to first cracking as a function of the corrosion rate and physical properties of the reinforced concrete, was never experimentally validated. This research study focuses on examining Bazant’s model and its validity using commercially available corrosion rate measuring devices. With validation of the model, the present condition of the structure could be assessed, damage rate predicted, and repair or rehabilitation methods selected.

The experimental program examined 18 test blocks and 52 test slabs constructed to simulate sections of a reinforced concrete bridge deck. The test blocks and slabs contained a single layer of electronically isolated steel reinforcement embedded below the surface of the specimens. The design variables included cover depth, bar spacing, bar diameter, and corrosion rate. Six chloride concentrations 0.0 to 9.6 lb/yd³ (0.0 to 5.7 kg/m³) added directly to the fresh concrete were used to vary the corrosion rate. Two environmental conditions were also considered: outdoors in Blacksburg, Virginia and indoors with near constant moisture and temperature. The slabs were designed with two reinforcing mats. The lower mat of fiberglass reinforcement was placed in both directions and the upper mat was one layer of fiberglass and one layer of steel. Thus, the corrosion rate a single layer of electrically isolated steel bars was monitored.

The corrosion rates were monitored from the date of casting for approximately one year using the 3LP and Gecor corrosion rate devices. The amount of metal loss of the bars was determined both destructively and non-destructively to allow a comparison between the two. The metal losses were compared to the predictions made according to Bazant’s equations as well as to other existing time to first cracking criteria to determine their validity. Tests were performed to verify the chloride contents of the slabs. The effect of temperature on the corrosion rate was also investigated. Preliminary results indicate that temperature strongly influences the corrosion rate.

To date no cracks have formed in the surface of the test specimens. The corrosion rate of the slabs has been progressing at a maximum rate of 1.24 mils/yr (31.5 microns/yr) for over one year. The test blocks and slabs are to be monitored until cracking occurs at which time Bazant’s model will be either validated or adjusted to reflect the predictions derived from the commercially available corrosion rate devices used in this experimental program. Additional tests must also be performed to further quantify the effects of temperature and other variables.